Connections

NEW YORK – Every media company today must innovate for the future, and analytics and machine learning (ML) in particular are being used by the most forward-thinking companies in the industry to maximize engagement throughout the audience lifecycle, according to Eric Schenk, technical director, office of the CTO at Google Cloud.

“The industry is undergoing these tectonic shifts right now,” he said Oct. 31 in the opening keynote “Insight-Driven Decision-Making for Media” at the NYC Television Week’s TV Data Summit.

“On the one hand, you’re seeing a lot of consolidation in the industry and, on the other hand, you’re seeing competition from new players that are delivering content to users in a more personalized sort of way,” he told the Summit.

The top issues for global media CEOs today are no longer the content itself or distribution, but instead changing audience habits, speed of technological change, and availability of key skills, according to Schenk, who pointed to the results of a PwC survey of such executives. Changing audience habits was cited by 81% of respondents and “that’s because it’s easy for customers to move from one platform to another in the new Internet age,” he said. That means it’s no longer the content that is “king,” but the audience, he told attendees.

Meanwhile, “technology is changing even faster” than audience habits, he said, adding: “With the advent of machine learning and virtualized cloud computing, everything is changing under your feet and how do you keep up with all of that?” He noted that 78% of media CEOs cited changing technology as the top issue they faced.

The third issue cited by media CEOs surveyed (75% of them) was availability of key skills that are needed as a result of all the technological advances, he said, calling that “even worse than” the other issues “in some sense because this pace of change requires special skills that you may not have.” Therefore, media companies must “find the people that know how to take advantage of this technology,” he pointed out.

As a result of all these trends, “innovation is not optional,” he said, adding: “If you want to not just survive, but thrive, you have to take on this change and really think about how [you are] going to do things differently. How are you going to keep up with your audiences? How are you going to make decisions about programming, content creation, segmentation? How are you going to think about personalization, monetization? These things are all difficult questions to face.”

Google’s take on it is that “in order to answer those questions, you have to harness your data,” he said, noting that data “cuts across” all aspects of a company, including content creation and distribution, audience engagement and monetization/modernization.

“If you want to use the data successfully,” he said it “has to be used in a two-way street that brings it all the way back into each one of these stages of your media pipeline and actually informs the decisions you’re making all the way through, even as far back as content creation.”

The problems created by data include the increasingly huge volumes of it that exist today, as well as how quickly it’s being created, and also data fragmentation, he went on to say, pointing out that data is often spread out across multiple “independent siloes scattered throughout your companies,” he said. Some of that data may date back to legacy systems from many years earlier, he noted.

The first step that media companies must take to “leverage your data to be able to drive better decisions is to consolidate it into one place,” he said, noting Google’s platform was designed to help customers accomplish that.

Once the data is all in one place, a company can then focus on data analytics that can yield improved results, he noted. Despite the growing interest seen in artificial intelligence (AI), he cautioned that companies shouldn’t just jump into AI without first strengthening their analytics capabilities and figuring out what questions are important for their specific businesses. Once an organization figures that out, then AI may be able to help, he said.
One of the goals of Google Cloud is to try and take the complexity out of ML for its customers, he noted.

He went on to say that data and ML are changing TV in terms of how content is created, how audiences engage with content, and the ongoing fight against content piracy.

But, “at this point, we’re just sort of getting started in this journey of how we might actually leverage the information that’s inherent in everything that we do in the media space from the beginning to the end with the customers,” he told attendees.

He left attendees with three key takeaways. First, companies must figure out what data they have and gather it all together in one place. Second, he encouraged organizations to “build a unified data analytic capability if you don’t have one already.” Finally, he said: “Remember this is early days. We don’t know the best way to leverage machine learning in the media space yet. It hasn’t been discovered. It’s still out there to be discovered by companies like yours. And the key to that is to find cheap ways to experiment. And you should also look at what’s happening in the adjacent media spaces,” including games and music.

Prior to Schenk’s keynote, Tim Hanlon, CEO and founder of media industry consulting and investment advisory firm The Vertere Group, called this a “very transformative time in television’s history.”

Making note, like Schenk, of the significant amount of rapid change that’s taking place in the sector, Hanlon said, “not everybody in the television industry necessarily is comfortable” with that.

Whether you are in the television space today and “grappling with all of this influx of digital data and what to do with it and what becomes of it, or you’re on the other side of it – you’re in the data space yourself or you’re in the technology space and you see the ability to be able to harness this stuff and actually transform media and television – both sides of that equation need to really come together,” he told attendees.