We are looking forward to attending CES one month from now to see “the new new thing.” Technology and innovation are enveloping our lives, and it will all be on display in Las Vegas.
One of the key trends that will permeate many of the exhibits at the biggest tech show on the planet is AI – or Artificial Intelligence. If you’ve been enduring HBO’s “Westworld” the past few months, you’re seeing AI in its scariest extreme, especially the ways in which human beings interact with robots or “hosts.”
In the real world of 2016, however, AI or robotic thinking is impacting many aspects of our daily lives – including the music we listen to.
Perhaps our first encounter with algorithms was Pandora. The formula that grouped music together was based on their Human Genome Project – a combination of human coding of songs mashed up with their recommendation engine.
Seemingly at the beginning, it worked. My personal experience with Pandora is that music discovery – especially new artist discovery – was much more common and joyful in the early days than during the past several years where the pure-play is often as predictable as most radio stations.
Now The Financial Times reports that AI has become a key cog in the Spotify selection process, especially their Discover Weekly Playlist. Their algorithm combines listening preferences with others consuming the same or similar music to make recommendations.
In the FT story, writer Nic Fildes reports that Google Play is taking a different approach to AI, factoring in smartphone data: time of day, location, and even movement to determine whether the user is active at the gym, bored in a dentist office waiting room, or working away on the job.
And London startup Quantone (“the most in-depth data in the music industry”) says they’re using IBM Watson to enhance music discovery by factoring in reviews, blogs, and social media comments to algorithmically select the best songs possible.
So if you’re a DJ staring at a screen of the next hour’s music, pre-scheduled by your program director, what hope do you have of effectively competing with these data-driven competitors?
While you may not have the flexibility to juggle songs or insert what you think would be appropriate, you may in fact have a more powerful tool: the local context and your knowledge of the music and why it matters to your audience.
Unlike Spotify, Pandora, Google Play, and even Zane Lowe, Apple Music’s talented DJ, you’re sitting right there in Pittsburgh, Poughkeepsie, or Panama City staring out a window (hopefully) at your city or town.
You know whether it’s raining, snowing, or brilliantly sunny. You know that if it’s 4pm, a large portion of your listeners are stuck in heavy traffic on the way home. You know whether your audience is heading to your town’s Christmas tree lighting ceremony or awaiting the big bowl game featuring one of your area teams. You know Twenty One Pilots tickets go on sale this Friday. And you know Aerosmith played your city’s biggest venue last December.
But all too often, listening to “local radio” is an exercise in generic entertainment. I am well aware of format restrictions that regiment talent, strictly limiting breaks to be in compliance with the format guidelines. But at most stations, there’s precious few seconds that can be used to localize and contextualize a break to make it resonant, relevant, and meaningful.
The original Top 40 icons of radio knew precisely how to pull this off – always over the beginnings and ends of two-minute songs. That’s what separated Dan Ingram, the Real Don Steele, John Landecker, and Scotty Regen from everyone lese. You always knew whether you were in New York, L.A., Chicago, or Detroit when you were listening to these guys.
So, do I know where I am when I’m listening to you? Is yours a show that could only sound great if I’m sitting in Phoenix, Des Moines, Norfolk, or wherever you’re broadcasting from?
As in programmatic buying, algorithmic streaming radio is a fact of modern-day life. In time, those robot DJs are going to be fine-tuned to read our inner thoughts and select the music that fits our mood of the moment.
And just as a great sales person can put together an integrated marketing campaign for a local client that utilizes the very best a brand has to offer, a local DJ who’s aware, awake, and alive can deliver a show that makes us want to keep the radio turned on.
Unless, of course, we’re convinced that turning our DJs into robots is the smartest move we can make.
Spotify, Google, and Amazon have superior AI than any radio broadcasting company can offer. And that’s why local radio needs to cultivate and grow its most unique strength…
Great local personalities.
It’s time to double down before the robots take over.
Jacobs Media has consistently walked the walk in the digital space, providing insights and guidance through its well-read national Techsurveys.
In 2008, jacapps was launched - a mobile apps company that has designed and built more than 1,300 apps for both the Apple and Android platforms. In 2013, the DASH Conference was created - a mashup of radio and automotive, designed to foster better understanding of the "connected car" and its impact.
Along with providing the creative and intellectual direction for the company, Fred consults many of Jacobs Media's commercial and public radio clients, in addition to media brands looking to thrive in the rapidly changing tech environment.
Fred was inducted into the Radio Hall of Fame in 2018.
- Great Radio – In The Niche Of Time? - November 20, 2024
- “Be My 9th Caller Right Now And Win A Free Slurpee!” - November 19, 2024
- Say Hello To The “MYdia” Generation - November 18, 2024
Amen! Personalities make radio great.
Thanks, Clark.
Right on! Keep it relevant, compelling and the place your listeners go to feel at home. For music, fun, info, weather…local.
Thanks for the comment, Bill, and for reading our blog.
Thanks for advocating for my job to continue to exist!
Although… I did wonder if the AI was personalizing this post for me when I read “…you’re sitting right there in Pittsburgh, Poughkeepsie, or Panama City staring out a window (hopefully) at your city or town.”
I am staring out a window in Pittsburgh!
Joey, the JacoBLOG has data that you can’t even imagine. Thanks for writing and I’m glad it resonated for you!
Yes, yes, yes…but don’t tell the other stations….
🙂
It comes down to money, in my experience, Fred. Lower overhead and less risk with tightly scripted robo-jocks or AI. Less chance of huge success – but less chance of failure and perhaps a smaller but more consistent revenue stream. Which takes the joy out of it for both the listener and the “personality”. What’s more, the human touch is based in imperfection and irrationality, and if we lose the human touch we might as well close up shop. AI doesn’t have value in creativity – radio or otherwise.
Dave, I agree with you about resources, but it also revolves around fear. Playing it super-safe can work – up to a point. Personality radio isn’t always pretty, but it’s the human element that provides the medium with its best chance at long-term survival. Thanks for the comment.
I agree with everything you wrote. I applaud you for saying it. The problem is our measurement system tells us to be more robotic. It tells us to play only the hits and rewards tight playlists with little or no personality. How do we overcome the robots in a PPM world?
Thanks for the kind note, Kayvon. No question that PPM encourages less interaction at a time when radio needs to step it up. Air talent needs to stay focused on delivering strong messages but in a timely basis. And I will say that some of the most successful stations feature personalities who are personalities. Robotics will not win in the long haul. Appreciate you weighing in.
Fred
I am honored that you thought enough of my work to include me in that list. That I’m included with The Real Don Steele blows my mind!!!
John, great to hear from you – always a fan. My money’s on you guys in the coming “Robot Wars.” Thanks for chiming in.
AI or not a music ‘collection’ is not the same as live radio, for many of the reasons you state. When the local station and it’s DJ’s make that emotional bond with listeners, magic happens, listeners become loyal fans.
Appreciate the comment, Brent. Robots don’t make magic.
Fred: I voice track an all night show on Dayton’s WHKO-FM (K-99.1). Not trying to toot my own horn here, but I have never believed that a voice tracked show “can’t” sound in the moment, local and live. On my pre-recorded show, I have:
1.) A twice a night visit from our TV meteorologist talking about what
weather is expected for later in the day. (and overnight when
possible.
2.) Send ups to local business open on the third shift.
3.) Music and artist news.
4.) A morning “oldie” feature sometimes driven by something on this
date in history, sometimes driven by an “open mic” request sent in
by a listener.
5.) Weather mentions hourly and more so in the final 4 AM hour (as
people in our town are getting up earlier to go to work.)
6.) Talk about upcoming events in town.
7.) Personal observations about events that happen to me to which I
think the listeners can relate. (such as: rushing out to get a last
minute gift on the morning of a family Christmas gathering having
realized I forgot somebody, and thanking the store that helped me
with the gift.)
8.) The usual station promotional mentions.
I would defy anyone who would say the show sounds canned.
Kevin, absolutely brilliant. It should take as much work to prep a voicetracked show as it does a live one. Kudos to you for bringing that approach to your show. That said, too many broadcasters view voicetracking as a convenience rather than as an opportunity to do a great show that just happens to be canned. Appreciate you telling your story and raising the bar.