In a world of data, where is the emotion? Over the last few years at Beet.TV, we have reported on several new approaches to ad targeting that ask that same question.

In one of the most notable, Unruly Media’s system indexes web video by emotion – like “exhilarating”, “sexy” and “cool” – so that advertisers can target responses, not people.

But one of the newest such methods comes from one of the biggest publishers in the world.

The New York Times has crowdsourced the emotions experienced by thousands of its readers, in response to the stories they read, and is now packaging up those emotions as a data point for sale in ad auctioning systems.

“What we’ve been doing with Demo, our initiative, is take some of the tools and data that we use for personalization or that we use in the newsroom to explore stories, and turn them into products for marketers,” says Allison Murphy, ad innovation VP at the Times.

“One of our recent ones we’ve called Project Feels. We built a data set of how people respond emotionally to stories. We asked thousands of readers to read thousands of articles and then said, ‘For each one, how did this make you feel?’

“Based on that data, our data scientists built a machine learning algorithm, so that now we can look at any piece of content and predict a range or an array of emotions that readers may feel. We are using that now for advertising.”

In these heady political-news days, you would be forgiven for feeling like readers would respond to Times stories with emotions like “disgust”, “revulsion” and “fear”.

But news doesn’t have to mean negativity, because publishers like the Times don’t just publish hard-hitting reporting – there is lifestyle content, too.

“We built a product, we call it Perspective Targeting. It’s a line item in DFP, and it allows a marketer to say, ‘I want my media to run next to content we think will evoke one of a handful of emotions, like self-confident or inspired’,” says Murphy.

“You can imagine a brand that has a really positive message for young women or for a particular consumer set choosing self-confidence. Their media is going to run against content that we’ve predicted will make readers feel self-confident or evoke that within them.”

This video is part of Beet.TV’s coverage of the Digital Content NewFronts 2018.  The series a co-presentation of Beet.TV and the IAB.   Please see additional videos from the series on this page.