How Coca-Cola targeted ads based on people’s Facebook, Instagram photoson May 16th, 2017
When Coca-Cola wanted to push iced-tea drinkers to consider its Gold Peak brand this summer, it didn’t target people like most brands do by using their search history. Instead, it combed through consumers’ photos on Facebook, Instagram and Twitter and served them ads based on images they shared on those platforms.
Gold Peak tapped into an image recognition engine that identified people who posted images that featured glasses or jugs of iced tea, displayed emotions such as happiness and excitement as well as contained cans or bottles of its competitors, including Snapple, Honest Tea, Lipton and others. Those people were then served Gold Peak ads on 40 mobile sites and apps after leaving Instagram, Facebook and Twitter.
For example, if you posted a picnic table spread with a jug of iced tea somewhere in the mix on Instagram, Gold Peak could have targeted you with ads while you read an article on Business Insider or checked the weather on the AccuWeather app in the past month, thanks to your photo.
“We’ve been using social listening for targeting for years, but people hardly use these social platforms to share text anymore,” said Benjamin Bring, vp of media at IPG’s Ansible. “Unlike text-based targeting, which can often be ambiguous, pictures provide more nuanced and subtle context.”
Cluep, a Toronto-based company, developed the proprietary image recognition engine. The engine learns from every publicly posted image it sees on social networks to identify brands, products, logos and scenarios. Then, it engages consumers around their interests, activities and lifestyles on mobile apps and mobile websites outside of the social networks.
Brands including Nike, McDonald’s, Toyota and Under Armour have partnered with Cluep for this type of visual targeting. While image recognition itself isn’t new, Karan Walia, Cluep CEO and co-founder, said Cluep was a step ahead of the platforms themselves. While platforms like Facebook have their own image recognition engines in place, they are not leveraging the technology to allow brands to target based on pictures consumers post. Furthermore, Cluep uses its own ad server and real-time bidder to serve ads outside the social media platforms and in premium mobile apps and mobile websites through its partnerships with SSPs. Cluep has partnerships that give it access to all public data on Twitter, Facebook and Instagram as well as partnerships with SSPs that can then result in automated ads for its clients.
“We have bridged the gap between image recognition and advertising at scale,” said Walia. “No ad tech platform is doing image-based targeting like we are.”
What’s more, people don’t seem to be creeped out. Brands that have run beta tests are seeing conversions and click-through rates of five to 10 times the industry benchmark on premium publishers, according to Walia. Coca-Cola itself has seen a click-through rate of over 2 percent, which is three to four times its usual benchmark for the Gold Peak creative.
“The biggest KPI for Coca-Cola is driving more people to the store, and consequently, store sales,” said Bring. “If we keep performing, we will certainly look at doing it for other clients.”
The post How Coca-Cola targeted ads based on people’s Facebook, Instagram photos appeared first on Digiday.
SEE ORIGINAL POST HERE