It’s impossible for fans to have too much information about their favorite artists — which presented a serious challenge to the Recording Academy’s editorial team as nearly 900 artists are nominated for Grammys in almost 100 categories. After partnering with IBM, the small editorial team was able to scale the volume of their work — serving up timely, relevant content to millions of fans across multiple channels.
It’s impossible for fans to have too much information about their favorite artists — which presented a serious challenge to the Recording Academy’s editorial team as nearly 900 artists are nominated for Grammys in almost 100 categories. After partnering with IBM, the small editorial team was able to scale the volume of their work — serving up timely, relevant content to millions of fans across multiple channels.
It’s the first Sunday in February, and a Los Angeles arena is packed for Music’s Biggest Night® – the Grammy Awards.
Millions of fans worldwide are turning in to see which of almost 900 artists will bring home the golden gramophones in close to 100 categories.
But as the show’s producer, you’ve only had three months since the nominations were announced to promote the best of music this year, across every genre and on every conceivable digital platform.
The National Academy of Recording Arts and Sciences faces that challenge every year.
Panos Panay: Attention is in short supply. But we depend on that as an organization. Not just for the awards, but ultimately for all the other activities that we're doing, that are there to celebrate and advocate for creators.
Capturing audiences is an important challenge that any organization has.
That’s Panos A. Panay, President of the Recording Academy. To address that challenge, the Academy partnered with IBM to create an AI-powered content engine that could quickly create social content and insights about every Grammy nominee during the compressed awards season.
From IBM and Bloomberg Media Studios, this is “The ROI Rules of AI,” and I’m your host, Edward Adams.
On this podcast, we’re exploring how organizations of all sizes are using AI to remake their operations, increasing their return on investment and that of their customers.
Panay graduated from Boston’s prestigious Berklee College of Music with a degree in music business management, and later served as the institution's Senior Vice President of Global Strategy & Innovation, overseeing its campuses in New York City, Valencia, Spain and Abu Dhabi. He’s also an amateur guitarist and trombonist.
The non-profit Recording Academy is composed of 22,000 members – not just performers but songwriters, arrangers, engineers, producers, and everyone else in the music creative community.
It advocates on public policy issues affecting music, undertakes educational activities, and supports musicians in need through its MusiCares philanthropy. But it’s best known for celebrating excellence at the annual Grammy Awards.
The Recording Academy has a five-person editorial team, but the need for content about Grammy nominees outstrips their ability to create it during the busy awards season. Having worked with IBM on a variety of projects stretching back almost a decade, the Academy saw this as an opportunity to use IBM’s watsonx technology as a force multiplier.
Panos Panay: I've always seen technology as a great amplifier of human creativity. So in collaborating with watsonx, we've been looking to amplify the capabilities of our own editorial team.
From our standpoint, every nominee, every genre, irrespective of whether it's popular or not, deserves its own moment in the sun, its own celebration. This is where our collaboration with watsonx can be extremely transformational in our ability to celebrate not only a handful but every single one of our nominees.
IBM and the Recording Academy began the two-month project using large language models, and then infused them with all of the content and data the Academy had created over the years about nominees.
That also changed the role of the Academy’s editorial team from that of reporters and writers, to becoming editors who checked every piece of content created for accuracy and the Grammy’s brand voice.
Panos Panay: Our people, they're absolutely hands-on. It's always great to have a human editor taking a look.
Content ranged from artist insights to enhance the GRAMMY Livestream viewing experience to social assets about various award categories and nominees.
Panos Panay: We generated over 4 million impressions across our social channels and over 1 million views on live.grammy.com during our livestream.
The project delivered the Recording Academy’s most valuable commodity – the attention of music fans, says Panay.
Panos Panay: The single most important currency today is attention. And ultimately, all of us -- whether you're a corporation that's selling a widget, or whether you are an organization like the Recording Academy -- the intelligent creation of content is a megaphone for your activities.
The Recording Academy had set itself up for success in two ways before ever undertaking the content engine project, according to Brian Fallon, Vice President of Data, AI & Automation for the US National Market at IBM.
Brian Fallon: One, they were maniacally focused on the quality of their data, right? That's the basis of everything, because what you're going to put into this is what you're gonna get out at the end. And they already had done that in a meticulous fashion.
Two, they had set a policy for how they were gonna use AI. And I would encourage every company to do this. We've done it inside IBM. You gotta understand what your guide rails are so that you know what you can solve with the technology, versus what you're going to require humans to do.
AI has the ability to unlock value from existing data that is in a variety of formats, Fallon says.
Brian Fallon: The data challenges that the Academy faced was not much different than other organizations. It's multimodal in terms of not just binary data. It's videos, it's locked in PDFs, presentations. It's across many forms. How do you unlock that, and unleash the power of that data to your internal constituents, and your external clients? That's really what we were able to do with this project.
Writing content about Grammy nominees is one of those…
Brian Fallon: …moments that matter. Moments that are highly repetitive but manual in nature. And frankly, ones that your internal and external clients complain about. Those typically are the use-cases where we wanna start, and then we work backwards to utilize the data, both known and unknown, internal to the organization and perhaps externally available, to help go solve those challenges, leveraging the power of generative AI and automation.
For the Recording Academy, the return on investment in their AI-powered content engine ultimately wasn’t just a means to create a lot of content, but a means to communicate with its audience on an emotional level, Panay says. It gave music fans a little taste of what he experiences every year behind the scenes at the awards ceremony.
Panos Panay: In terms of being in the room during the Grammys, it's one of the most amazing experiences, frankly, you'll ever have. I've been in the music industry for 30 years, but there's not been a Grammy night that I've attended that didn't blow me away. Even sometimes among the most celebrated stars, getting a Grammy and what it means to them, it's really touching.
On your obituary, if you've won a Grammy, it will be likely either in the title or on the first line. So, being in the room when somebody's life truly changes, that specialness doesn't go away. Ultimately, music is an emotional thing. And for us, connecting with audiences around that is extremely important.
The program with watsonx and the Recording Academy captures everything we want to be and the stories we want to develop as a means of creating that emotional engagement.
This has been “The ROI Rules of AI,” a podcast from IBM and Bloomberg Media Studios.
If you like what you hear, subscribe and leave us a review.
I’m Edward Adams, thanks for listening.