Tamr: Machine Learning Can Be Used to Transform Creative Talent Management – Media & Entertainment Services Alliance M&E Daily Newsletter

Machine learning can be used by the best talent managers today to transform creative talent management and find the right opportunities for their clients, according to Matt Holzapfel, solutions lead at enterprise data unification and data mastering specialist Tamr.

In an industry that runs on storytelling, its stories are increasingly informed by huge amounts of data: hundreds of datasets, millions of records and billions of data points (including tweets) from sources inside and outside the business. By using machine learning to serve up analytics-ready data from disparate data, creative talent management firms can create very human stories with mutually successful outcomes for clients and media companies time and time again.

Tamr helps large organizations clean up dirty data so that they can get that data ready for their analytic and digital transformation aspirations, Holzapfel said during a May 27 presentation at the Hollywood Innovation and Transformation Summit (HITS) Liveevent.

During the presentation Using Machine Learning to Transform Creative Talent Management, he explained how Tamr helped Creative Artists Agency specifically use machine learning to take a new lens to what the data management ecosystem should look like in order to transform how they were using data and analytics within the company.

In the process, Tamr was able to dramatically increase the throughput of their analytics and help drive more insight for their agents, he said.

Within every industry, the old saying is your biggest assets leave in the elevator every night, he noted, adding: Within entertainment, nothing is more true in that people are the entertainment industrys biggest asset. The actors, the musicians, the artists that people pay to see [are] really at the heart of the entertainment industry.

And he pointed out that one of the biggest challenges within the industry is how you match the right talents, the right piece of content for the right audience.

It is often not the end analytic that is the most challenging part, he told viewers, explaining: I think in a lot of cases, when were talking about data, were usually thinking about those analytics: the visualization, the model whatever it is that comes out the other end that helps us make a decision. However, what often is the biggest bottleneck is the data around it, he said.

As an example, he noted that we can look at actor Vin Diesel and try to gauge his social reach, the top demographics that include his fans and what an ideal role for him would be where a company could attract a big audience and be successful.

If the data is readily available at our fingertips and nicely organized, then these questions become pretty quick to answer, he said, adding: We can answer these questions in seconds. But often today they take weeks [to answer] because the data itself is not neatly organized. If we want to understand who is Vin Diesels target market [and] what roles should we put him in, that involves pulling audience data, YouTube data, social media data about what are people talking about [and] what the sentiment is like.

Some of that data is structured and some of it is unstructured, he noted. But the bottom line is that its extremely buried and scattered everywhere and so it makes it difficult to even have the information needed in order to make decisions confidently, he told viewers.

At the end of the day, any decision within this industry is a bit of a leap of faith, but without the data to back it up, youre often just kind of flying blind, he said.

Once you get the data organized in a warehouse, the next problem that companies face is the data itself is dirty, he noted.

If you want to figure out the impact of, for example, Steve Carell on the TV show The Office, you have to sift through all of this data, and just wrangling and organizing all this data is often the bottleneck for such analytics, he said.

That was a key part of the bottleneck at CAA no matter how much data they were acquiring, they were just running into more and more issues with actually making the data usable, he told viewers.

However, the good news is that, particularly over the past handful or so years, the tools that are available the solutions in the market have evolved quite a bit and we now have what we need in order to solve this problem, he stressed.

Traditionally, the way the market has looked at this problem has been kind of twofold: You have your source system in which you just collect all the data you need so everything is in a warehouse or data lake, and then you need people who can analyze all that data and figure it out, he noted.

The problem with that, however, is many of those analysts, who are very scarce and difficult to come by, end up spending a lot of their time doing one-off cleanup and data preparation, and not on analytics, he pointed out. These kinds of human-intensive approaches are difficult to maintain and lead to poor productivity, he said.

However, what used to take weeks to gain insight now takes only minutes because companies are starting to see their data as an asset and are focused on the data engineering, enabling the prep to be done upstream, he told viewers. That is dramatically reducing the amount of time analysts and data scientists are spending preparing and getting the data right, he said.

And CAA is one of the best examples that weve seen in the media and entertainment industry of reducing the time to insight from two weeks down to two seconds, he noted.

He went on to stress: There isnt one silver bullet to solving this problem. There isnt a single suite or a single solution that you can buy thats going to do everything that you need to do in order to solve this problem.

Fortunately, CAA recognized early that it would need to invest in next-generation tools that are open and interoperable and enable you to have that agility to do it, he said. Also important was its shift to modern, cloud-based tools, he said, adding CAA took a completely cloud-first approach to the challenge.

Clickherefor the presentation slide deck.

The May 27 HITS Live event tackled the quickly shifting IT needs of studios, networks and media service providers, along with how M&E vendors are stepping up to meet those needs. The all-live, virtual, global conference allowed for real-time Q&A, one-on-one chats with other attendees, and more.

HITS Live was presented by Microsoft Azure, with sponsorship by RSG Media, Signiant, Tape Ark, Whip Media Group, Zendesk, Eluvio, Sony, Avanade, 5th Kind, Tamr, EIDR and the Trusted Partner Network (TPN). The event is produced by the Media & Entertainment Services Alliance (MESA) and the Hollywood IT Society (HITS), in association with the Content Delivery & Security Association (CDSA) and the Smart Content Council.

For more information, clickhere.

View original post here:
Tamr: Machine Learning Can Be Used to Transform Creative Talent Management - Media & Entertainment Services Alliance M&E Daily Newsletter