Lemon&Orange is helping Marks&Spencer to review marketing effectiveness of its shop windows. Using a unique technology, Lemon&Orange are supporting the process by providing statistical data regarding the number of glances on particular elements of the exposition. The software – Metrics Anylzer – recognizes faces of people passing by and assigns them to the appropriate age group and sex, transferring the gathered data to the analytical module.

This allows to measure demographic structure of passers-by in real time, i.e. the collected data are available seconds after the measurement.

The solution used by Lemon&Orange provides the statistics regarding the number of glances and allows to test their effect (eg. how a glance at a particulat object in the exposition correlates with the number of customers entering the store.)

Gathered data:

  • Total time in the front of the camera
  • Total time spent facing toward the camera
  • Pedestrian traffic near the camera
  • Hourly popularity of the ad


  • 91%  detection accuracy
  • 88%  gender accuracy
  • 78%  age accuracy


  • Software has no ability to recognize or identify anyone. It gathers purely numerical data, no personally identifiable information is collected and no images are ever saved

Main capabilities:

  • It allows you to measure the popularity of existing storefront (or to compare the popularity of a number of them)
  • It measures physical spaces such as retail stores, end caps, vending machines, door way entrances and cash registers