I’m like a broken record in many ways (shrill, useless, etc) but one of my most frequently spouted opinions is that the future of audio is i) live, and ii) social.
This opinion is becoming tougher to sustain: this week Spotify announced that it would be shutting down its Spotify Live service, previously called Spotify Greenroom (and described by PodNews as a “Clubhouse clone”). “After a period of experimentation and learnings around how Spotify users interact with live audio, we’ve made the decision to sunset the Spotify Live app,” a Spotify statement said.
Putting aside the hilarity of using “sunset” as a verb meaning to “kill”, this is part of a trend against live and social audio. Reddit Talk was sunsetted out of its misery last month, and Facebook’s live audio rooms were brutally sunsetted back in December. Clubhouse, the app which briefly sparked a live audio obsession, was valued at $4bn in April 2021 – it is now valued at a carton of cigarettes and a sharpened stick.
So, was I wrong?
I am frequently wrong about things. The prime example of this is newsletters, which I always felt were rubbish. You couldn’t accurately measure consumption, you carried huge chaff in your mailing list, you were subject to strenuous data requirements like GDPR, and fundamentally you were outputting the most passive of all content. It felt like weak tea compared to podcasting’s strong coffee. And yet, the years since I started voicing that opinion have proved me wrong. Publications have invested more in newsletters than podcasts (on the whole), and have received better returns. I am, of course, writing this in a newsletter, and that’s all because, for independent publishers, it’s 10x easier to generate revenue via a newsletter than via the audio equivalent.
Some of my concerns about newsletters hold up, of course, but there are plenty of concerns about every medium. Pobody’s nerfect, and all that. This is all just to illustrate the fact that I’ve been wrong before and, when proved wrong, am happy to admit it.
I still don’t think I’m wrong about live audio. I still think that if I were a proper, gutsy new media entrepreneur, live audio is where I would be raising and spending money. The problem, for me, has been rooted in the philosophy of how these organisations consider their live audio projects. To better understand how live audio can work, going forward, you have to first understand how unexciting it is. This is not a revolution; it’s a new take on the wheel.
Things that are live audio: radio (obviously), Twitch, Zoom, Skype, phone calls, bedtime stories with your kid, jazz clubs, slam poetry nights, FaceTime, two plastic cups linked by a piece of string, a long pipe, Discord. Etc.
Take the latter example, because I think it’s the most important here. Discord is currently the most effective user of live audio amongst all major tech services. Why is that? Because Discord understood this key message: digital live audio is more effective as a social format than a broadcast format. The impulse was clear: in the 1990s, the birth of the video game generation, kids used to play GoldenEye with their landline pressed to their face. They wanted to have the sense that they were playing with their friends, even though this was pre-long-distance gaming. And so the phone became a conduit for that, a facilitator of a community.
When Discord launched in 2015, its founder Jason Citron explicitly called it a voice messaging service for video games users. He had suffered, as a League of Legends player, from an inability to communicate tactics with his teammates, and so Discord was created to bridge that gap. Even though it is still heavily focused on video games (not always to its merit), Discord has expanded beyond that original purview. It now has a valuation of $15bn and over 150mn actively monthly users (bear in mind Twitter, which sold for $43bn, has 450mn monthly users). And part of this rapid expansion of its userbase is because it never marketed itself as a broadcast mechanism.
What is the difference between social and broadcast audio? The former could be perceived as a way of digital audio replicating the feeling of being part of a group chat, or a conversation amongst friends, while the latter is simulating the feeling of being a radio presenter. And the natural impulse amongst start-ups in this space has been to try and push the latter: bring in experts, celebrities, genuine presenters, and have them use the format like they would a breakfast show on commercial radio. It would’ve been strange if Spotify, whose audio strategy has been very celebrity-led, had done anything else.
“We believe there is a future for live fan-creator interactions in the Spotify ecosystem,” the Spotify statement on its live audio future went on to say. “However, based on our learnings, it no longer makes sense as a standalone app.” These “learnings” touch upon two key issues. Firstly, it is a recognition that the strength of live audio is in that interactivity, that stepping away from a purely broadcast function. And secondly, they have learnt that these functions need to be better integrated. If you cannot provide a service within your core app, your core functionality, you might as well not provide it at all.
There is a point where social and broadcast audio meet, and that’s the sweet spot that hasn’t been fully exploited yet. It has been exploited (though not to its full potential) with video streaming. Think about the community focus of streamers on Twitch or YouTube. There is a sense that the people broadcasting via these channels are celebrities; the disc jockeys of the digital era. But they are also fundamentally accessible. Talking to them is hard-coded into the success of the format – pay $5 and have your message read out on screen. Pay $20 a month, and maybe they’ll shout you out. Just go on one of these channels – any of them really – and see how often the host directly addresses “chat”. They’re all doing it, because they all understand that their success is based off a perceived conversation.
And conversation is the key. Clubhouse fizzled, in my mind, when people realised that it was predominantly being used by a small group of elite (to use the Trumpian term) users, who were shilling crypto or wellness or generic lifestyle advice, and expecting a silent audience in return. There were definitely other issues around the low audio quality, and the generally unprofessional way in which conversations were broadcast (as has been a perennial problem with live audio, and another area where there should be a concerted effort to catch-up with the professionalisation witnessed in streaming video) but, for my money, the issue was the sense that things were monodirectional. Clubhouse utilised a brilliant marketing trick – the Golden Ticket invite system – to attract its first, core userbase. But then it immediately disenfranchised 80% of them when they realised that it would be, for them, purely a listening app. People want to listen, sure, but they also want to be heard.
That’s why I think it’s vital to return the social element to these apps. The Spotify spokesperson has it spot on: Spotify Live, for the first time within the Swedish audio giant’s premises, gave users the opportunity to interact with creators. That’s an incredibly valuable proposition, financially, if harnessed correctly.
I suspect that Spotify may simply not be the right company to exploit live audio. I’ve written before about how their corporate strategy seems to be to become an everything-channel for sound (YouTube for audio, to put it succinctly) but I also believe that Spotify will always be seen as, first and foremost, a music streaming platform. Drake and Ed Sheeran and Dua Lipa will always be the faces that spring to mind when someone mentions Spotify. And opening fan-creator interactions at that level would be a fruitless deluge, where freaks shriek lewd comments at Rihanna, or hawk the latest memecoin in Harry Styles’ chat.
The perfect balance, for live audio, would be cultivating a space, not for mega celebrities to endure awkward interactions with fans, but for semi-social audio, broadcast to 10,000+ listeners (or whom 1-5% might be paying subscribers). That means you’ve got 100 to 500 paying subs on a channel at a time (ka-ching) and 10,000 people hearing adverts. That’s a sustainable revenue model straight away.
It's also unmistakeably Twitch. Where audio can, I think, diverge from just replicating streaming video (a divergence that is probably optimal, but not necessary because nobody has made a functional live audio app that has the same rough blueprint as streaming video) is by being more social. Streaming video is still dominated by single-person, to-camera content; streaming audio can, and should, be more expansive. The opportunity to hear a group chat is there and we already know people love it, because people love podcasts. This is not a space that should be dominated by monologues. Audio-only cannot rely on visuals to fill in blanks. But it can be hugely social. Listening parties, watch parties, live sports commentary, video games commentary, political analysis, dating advice, horoscope reading: all of these things would fit perfectly within the remit of a more social live audio.
The question is who will build this functionality, especially when places like Spotify, Reddit and Facebook are sunsetting their in-house apps, and Clubhouse has withered on the vine (too little sunlight, I suspect). The experiment with Clubhouse also exhibited how hard it is for insurgent publishers to generate, and sustain, an audience. And without an audience, live audio is just screaming into the void. So my biggest question about whether the live audio revolution happens is not whether it has a market (it does) but whether anyone is going to take a risk on building it.
The answer would, of course, be for Apple to build it into their core app suite. But beyond that, I do think that Twitter could easily –in five years’ time – be seen as a platform that has a 50:50 balance between text and audio. And perhaps Spotify can integrate social audio and chat functionality into its, very expensive, podcast slate. Maybe it’ll be TikTok, if it stays unbanned long enough. But, most of all, I wonder whether Amazon, who own Twitch (and Audible), might want to explore this area – they have the cross-industry clout to do it, and could start the revolution, as many modern revolutions do, with the captive audience of gamers on Twitch.
Or perhaps someone new does come from behind the pack (or behind the starting line) and enter this area. We are at a key moment for social media reinvention, with Twitter mired in controversy, Meta haemorrhaging money (and staff), and TikTok facing company-ending sanctions. It’s easier, right now, to see a social start-up catching fire than it has been at any moment in the last decade.
Keep reading with a 7-day free trial
Subscribe to Future Proof to keep reading this post and get 7 days of free access to the full post archives.