PRE

PRE can model and predict human engagement with complex visual content (such as web screenshots, in-store displays, videos, etc.).

These predictions can be used for A/B testing designs before implementation, measuring user flows across multiple web pages, measuring engagement with video, or, across omni-channel user journeys (e.g. in-store ⇨ mobile ⇨ desktop-purchase). PRE can also predict user flow in-store, which can help optimize in-store layouts/designs.

A/B test design options based only on a screenshot

Instead of spending significant time and cost implementing an idea, get a sense of how well it will perform within seconds using only a screenshot/design/mockup.

Predict multi-step user flows

PRE can predict the overall likelihood of specific user journeys across multiple steps (such as finding a service, interacting with the service, and making a post service action). This way, you can optimize the UX with specific user flows in mind.

Predict product engagement in video ad or social post

With PRE, you can get a quantitative measure of exposure of a product or service in a video segment. For example, you can predict how much engagement you are likely to get from an influencer video post, or, from a video ad.

Deep contextual understanding of visuals

Context matters. PRE analyzes complex images such as screenshots by understanding each element based on visual affinity and color. But it also analyzes context, including what is being featured and how engaging that might be to users.

Measure real-time engagement across retail channels

Predict engagement stats for any brand/product (even competitors) across all retail channels:

sephora

amazon

ulta

target

asos

walmart

feelunique

cult_beauty

* These are visually predicted percentages of product/brand engagement for users who land specifically on the lipstick listing pages of each of the above retailers. Actual engagements may vary.