Startup Analyzes Faces In Digital OOH Crowds To Measure How Response To Ads
May 18, 2015 by Dave Haynes
A San Diego start-up called Emotient is the latest analytics company to try getting some traction for face pattern recognition technology in the digital out of home advertising industry.
We’ve seen several companies come and go marketing products that use machine learning and cameras to look for, detect, count and roughly analyze faces that happen to look in the direction of screens. Attempts to get widespread adoption of using this tech to serve ads based on gender, broad age groups and even emotion have not really gone too far. It’s not enough to get Digital OOH on the plan, never mind introducing that kind of complexity to the buy.
Emotient is going at it a bit differently. Instead of real-time analysis of audiences, it is suggesting its face pattern technology can run reports on how crowds in places like bars respond to ads.
The company, on its corporate blog, says the web service, Emotient Analytics, analyzes videos of audience reacting to (or ignoring) content on these screens. It takes measurement beyond the “eyeballs within range” data available from current methods (e.g. phone location data) to the things that truly matter to advertisers – attention, engagement and sentiment. It quantifies these KPIs using facial position and expression, and does so on a time-stamped basis, so it can demonstrate the peaks and valleys of the true currency of advertising, human engagement.
To avoid privacy concerns, the company told Ad Age it does “anonymized aggregate analysis” and typically studies public venues where people are already being recorded on things like surveillance cameras.
“We keep the metadata,” CEO Ken Denman told an Ad Age writer. “We discard the image.”
Emotient recently worked with an undisclosed NBA team to analyze how faces in the crowd reacted to in-arena ads, including videos played on giant scoreboards. Theoretically, the data could be used to determine when during a game certain content gets the most attention and best reaction.
By analyzing faces in the crowd, Emotient also revealed a surprisingly basic fact: More women were attending games than the team had realized. The team “had no idea,” Mr. Denman said. That finding could result in more ads from brands targeting women, he said.