The Drum Awards for Marketing - Extended Deadline

-d -h -min -sec

Microsoft Technology Content

Ooyala forges partnership with Microsoft to score and analyze content for brand use

Author

By Laurie Fullerton | Freelance Writer

May 2, 2017 | 3 min read

While there’s lots of people populating video content, it takes too many others to screen and label it. That’s where artificial intelligence can provide a big assist.

While “it’s early days (in such use of AI) it’s an exciting solution to bring to market,” says Scott Braley, GM, Advertising Platforms, Ooyala, which recently announced a partnership with Microsoft’s machine learning platform.

In this interview with Beet.TV at the 2017 NAB Show, Braley outlines the potential benefits of machine learning for helping advertisers steer clear of unsafe environments. “The biggest challenge is that people have to sit and watch the content to see just what is and what isn’t appropriate. User generated content is difficult to label. This partnership allows us to score and analyze content more readily, without having to deal with armies of people reviewing content.”

There is some labeling and metadata associated with individual pieces of content, but the process usually doesn’t go far enough, according to Braley. “Certainly when you get into user-generated content, it’s very difficult to get all of that labeled properly,” he says.

AI can be deployed to score, label and categorize content “at scale and with more efficiency.” This would pave the way for more content available programmatically which “we believe will enable a more systemic connection between the sell side and buy side.”

As more consumers opt for alternatives to pay subscription TV models, younger audiences in particular are flocking to OTT viewing. With the shift in TV consumption, the younger audiences have been raised with the desire for choice. With the OTT becoming a great hybrid of the TV and digital world, deploying both together in a more practical environment is key.

Combining the best practices of the digital and television worlds would mean not deploying the hyper targetability and addressability of digital and eschewing “index-based measurement models like Nielsen for TV.”

Braley says this approach will involve targeting on a household level using as much census-level data as possible and “decisioning dynamically but maybe not with the one-to-one, user-base that we’re doing in digital today.”

Microsoft Technology Content

More from Microsoft

View all

Trending

Industry insights

View all
Add your own content +