Actual Audience Measurement

If you really want to make an impression, make sure your message can be seen. If no one sees it, no one wants to pay for it. We bring in a frugal approach to grab eye balls. We deliver real-time audience analytics so you know exactly the performances of your campaigns. Walkert AI’s computer-vision technology measures viewability, views and attention time, which elevate DOOH to the most accountable digital medium available today. Our data & analytics provide a unique understanding of a campaign’s performance as well as granular actionable insights to optimize a planning strategy, inspired by Quividi.

We follow an audience measurement process, lets take a look at it.

Viewing zone vs Detection zone

The viewability zone is part of the definition of what makes an audience impression. Walkert AI’s detection zone, in contrast, is based on the limits of detection technology (for instance, most cameras have a horizontal field of view of less than 70°). They are different metrics. It is possible, for example, for an audience impression to be valid and yet beyond detection.

The adjustment factor

To compensate for differences between the detection zone and viewability zone, Walkert AI’s detection metric, watchers, is multiplied by an adjustment factor to convert it into audience impressions. The mathematical model behind this adjustment factor is based on journey mapping.

The brown shade in the picture is the image take by the camera, suggesting there are 3 watchers. There are two more watchers but they are not in the viewabilty zone as the camera doesn’t capture them and are in blue zone. The third watcher is in the zone but is too far and hence is not a watcher. Hence, the adjusting factor.

Audience impression=Detected watchers*Adjustment factor

Journey mapping conversion

Walkert AI’s conversion model adjusts its detection metric (Watchers) in real-time into the human-valid audience impression metric that best evaluates the total actual audience of a display. This conversion model takes into account the characteristics of the display and of the camera (of the mobile device), as well as the field of view of each watcher. Walkert AI has developed a sophisticated journey mapping simulating engine, that adapts to every venue and simulates the traffic that would pass by the display and will be detected in the camera detection zone. The resulting journey mapping and detection data are compared to create the conversion model. The conversion adjustment factor is calculated both for the audience number, as well as their dwell time and attention time.

Real time conversion

Due to the nature of the modelling, it is possible for Walkert AI to provide live reporting of cross-platform human-valid audience impressions in real-time, despite any limitations in detection technology.

Audience measurement needs adjustments due to natural differences between the detection technology and the reality of the audience for the display. Hence, Walkert AI has incorporated an adjustment factor that gets an accurate audience measurement using an algoritm that is derived by our data scientists.