Polish DOOH Network Sees Traffic, Engagement Bumps By Using Computer Vision To Trigger Ads

A Polish media company is touting an interesting use of triggered content from AI-driven audience analytics that boosted foot traffic and social media engagement for a special event at a Warsaw mall.

By triggering targeted content only when pedestrians outside the mall were waiting for a crossing signal at a specific intersection, and when motorists were stopped at a nearby intersection, a campaign for an exhibition with full-sized superheroes at the mall saw foot traffic increase by 12% and spurred hundreds of social media posts, like selfies.

The campaign was run on a big transparent mesh LED wall that faces the nearby roadway for both walkers and drivers. It was put together by Screen Network – the biggest digital out-of home company in Poland, with more than 20,000 screens and 1,000 locations.

The technical side – the object and people detection, as well as the triggering – was done by the company’s innovations unit, Think DOOH. The pattern detection thing – picking up motor vehicles and pedestrians in defined areas – sounds futuristic, but is readily available tech these days.

The company has also done some novel content triggering based on external data or sensors:

  • Raising discounts on a Burger King milkshake promotion as daytime temperatures soared;
  • Running a promotion for a first responders-focused TV show when an audio sensor detected a siren in the area (yes, sensors can do that … there also ones for gunshots).

Lots to like about this approach, though it is one thing to schedule and execute these on a single screen than on a scaled network.

1 thought on “Polish DOOH Network Sees Traffic, Engagement Bumps By Using Computer Vision To Trigger Ads”

  1. Like you said, all the tech to trigger these things exists, it’s just a matter of someone making the decision to do something. A number of security cameras have software that can detect certain items withing defined zones and then trigger an alert. They can tell the difference between a person and a car etc.
    As for the temperature thing; I remember talking about this stuff when I got into digital signage back in the mid-2000’s. Old news but nice to see someone actually doing it.

Comments are closed.