“Music is the universal language of mankind.” — H.W. Longfellow
In December of last year I attended Google Play’s first startup event hosted in NYC. I was invited because my company, Cymbal, had the honor of being highlighted at Google I/O 2016, and then featured in a number of collections in the Play Store. I went alone and did not know anyone there beyond the hosts, who had their hands full with their hosting duties. I was on my own.
I was standing by the hot plates, fiddling with my phone and trying to muster the confidence to start a conversation with a stranger, when someone mercifully came over and, with enviable social confidence, introduced herself to me. Her name is Valerie Miller. I didn’t know who she was, and her humble demeanor didn’t create the impression that I might. But as I began asking her about herself, I realized, in a way, everyone already does.
In 2000, Valerie was a founding member of a company called MobileSpring. MobileSpring had a product called “Metcalf Inter-Carrier Messaging” which allowed cellular service providers to send texts and other content to other cellular service providers seamlessly. I listened to her, stunned, as she told me how her company engineered the foundation of a part of my life I take wholly for granted, all day every day.
Listening to Valerie, I was shocked to find that there was a time where such an obvious problem — you were not able to freely text whomever you like — was the norm. The choice of cellular provider is utterly irrelevant to an individual’s desire to communicate with someone. Valerie’s team understood this and fixed it. Today, about 20 billion texts are sent every day. She built a bridge, and now no one thinks about the bridge. We just take the journey.
As long as there has been music, we have shared it. Music means nothing in a vacuum: It has to make contact with you to have any meaning, and once it does, it leaves an indelible mark. If it comes from someone who matters to you, you’ll never hear it without thinking of them. If it is shared between friends, it defines the moment and changes your experience in real time.
But everything changed with the rise of Napster almost 18 years ago. Music was transformed from commodity to information and rewrote the rules of the game. Everyone learned that controlling the acquisition of physical goods is a very different problem from controlling the exchange of information. Sharing meant something new.
Suddenly, music felt the full effects of the largest unregulated market in the history of mankind: The internet. Exchange became instantaneous and limitless. This was a good thing because it made its delivery mechanism direct, global, and virtually effortless. But for many years it posed an existential threat to the world of people who made their living from music. In litigious moments, it posed a threat to listeners and sharers, too.
Paid streaming services changed all of this. The paradigm of music piracy could not only be embraced, but improved upon. Tamed. It was a really radical idea at the time: I’m in awe today of the hubris and conviction of the architects of streaming music, who marched up to the music industry, still-swooning and swinging from its sucker punch, to say, “We can fix your piracy problem and get people to pay you for it by beating them at their own game.”
There are now well over one hundred million people who pay for streaming music, and this number grows steeply year-over-year. While doubt about its business model persists — payouts are still unsatisfactory — the endpoint seems inevitable. Would you rather pay $10 for one album, or $10 for the entire history of recorded music for a month? Despite widespread skepticism about the model’s fairness to producers, a streaming subscriber pays twice what the average American music consumer does, and, at $120/year, almost 70% more than the average music consumer has ever paid annually, even at its peak in 1999. On top of this, it is now the #1 way Americans listen to music. It seems just a matter of time before the financial scales tip.
Listeners have more access to music than they ever have, and artists will be paid better than they ever have, too. But this sprint to transform the music industry to a new sustainability has created problems as it has solved them.
The thing that makes streaming music work is that, in one way or another, listeners pay to listen. Maybe that means hearing ads, or being gradually guided to a paid offering from a free one. But eventually, in some form, we all pay, and we can’t listen freely until we do. As this idea gained currency — both as a business model and an attraction — options proliferated. Spotify and Deezer blazed trails in the space, and soon bigger players in technology realized streaming music could not only be a meaningful business model, but an attraction to a larger ecosystem or operating system too. Apple Music became a powerful feature of iOS; Google Play Music beefed up Android; Prime Music helped sweeten Amazon’s amazing Prime offering for all 65m of its subscribers. As this technology became more sophisticated, it began to absorb adjacent offerings, like internet radio. Soon Pandora, a huge and venerable figure in the industry, launched its own on-demand offering to compete.
Today, the streaming music space is highly fractured. At first blush, one might assume that a single player will ultimately win the space, clearing lesser competitors out. But the institutional players that follow Spotify in subscriber hegemony — Google, Apple, and Amazon — enjoy structural advantages like coming pre-loaded to your phone, already knowing your (or your parents’) payment information, or having alternative revenue streams that can support a loss leader. Because streaming is not only racing to ascendance as music’s established form of delivery, but also becoming an attractive feature of larger companies’ offerings, it is reasonable to conclude that the landscape of streaming will be, for the foreseeable future, divided.
Streaming music’s fracture is compounded by its diminishing slate of competitive differentiators. Where these companies once thrived on disparities between their catalogues, all streaming libraries are trending toward parity as prized holdouts like the Beatles, Prince, and Taylor Swift fall. As streaming grows as the dominant means of consuming music, it becomes increasingly costly for artists to market their songs to anything but their greatest addressable market, which can only be found by listing songs on every streaming service.
This impending parity created an opportunity for cash-rich streaming services to create artificial differentiation with exclusive releases. This strategy was effective as a growth hack for a time, but it reached a peak boiling point last summer after Frank Ocean’s release of Blonde ended Universal Music’s employment of the practice and reignited rampant piracy. Now Apple Music is following suit by phasing out the practice. The exclusivity of Ocean’s album was a harsh reminder of the stakes of streaming music. Streaming succeeds because of the unique experiential advantages of its cultivated on-demand product over torrenting a la carte. Create unnecessary obstacles and piracy will once again rear its ugly head. The revival of the music business critically relies on this steady course of experience and access.
This brings us back to Napster. While Napster was not explicitly social, it architected what social music would mean for the internet with its single, de-commodified means of connecting, sharing, and listening. But times have changed since then. What is social music now? Is it even feasible?
To be social in any manner, it must be possible to find a common means to exchange or experience information together. Social interaction at baseline is predicated on finding a shared means of communicating and then engaging. Social music before the internet meant creating a mixtape for a friend, or listening to a song together in a car, or seeing a show. It meant being in the shared presence of sound, in body or spirit.
As the world sprints toward its splintered, streaming future, sharing music is largely impossible. The way we share songs now means using streaming links as our vessel, and because this model is fractured and paywalled, we can not guarantee our friends, followers, listeners, fans, and subscribers can listen to the link we are sharing. In fact, it is fair to conclude that, no matter what streaming service you share a song from, the majority of your recipients will not even be able to listen to the song you shared because they are not likely to subscribe the same service you do. They will tap your link and hear a preview, even though they 1. likely subscribe to a streaming service that 2. almost certainly has the song you shared available on it.
This is silly. Imagine if you couldn’t text a friend because she decided to sign up for a cellular plan through Verizon and you chose AT&T. If that blockage existed it would prevent texting from being an essential part of your life, like it now is for billions of people.
Streaming music is amazing, and it is the future. It is also an undifferentiated, restrictive, and increasingly unavoidable plurality. Interchange is restricted by consumer choices that are utterly irrelevant to an individual’s desire to communicate with someone. Someone has to understand this and fix it.
We launched Cymbal in May, 2015 on a simple, universal premise: Songs mean more when they come from friends, so let’s build a social network for sharing songs together. We initially built the app on the backs of the SoundCloud and Spotify APIs, giving our community access to an enormous composite library to share from. But after launching we encountered that same foundational problem for streaming music: Many of our users did not subscribe to Spotify, so they could not listen to the songs friends, favorite artists, and likeminded listeners were sharing, even though they had paid for those exact same songs on their own streaming services.
When Apple Music launched its streaming API, we were excited at the opportunity to allow more of our users to get full playback for songs, but wary of compounding this playback problem by flooding feeds with more songs that only a fraction of listeners could stream. Wouldn’t it suck to encounter a song in Cymbal available on Spotify, but only hear a snippet because a friend had posted it from Apple Music? We did not want to make our experience feel like a crapshoot.
So we dreamed up a crazy solution: What if we matched the whole library of Spotify to the whole library of Apple Music? Users already link their chosen streaming service to Cymbal in order to get full playback; If we know what streaming service they use, can not we just play the song they want to hear from the service they already pay for?
This was a eureka moment for Cymbal. We realized that this simple efficiency problem — getting a listenable link to listeners from any streaming provider — could be resolved for the whole internet too. Can most of your Twitter followers stream your Apple Music link? Do most of the subscribers to a subreddit have a Spotify subscription? You are a Deezer subscriber, but how many of your friends can you text a new song to? The numbers are against you.
What if we made the entire internet listenable, just by matching the right song to the right listener through the right streaming service?
Say you are a music publication. Your objective is to share songs to your readers, right? Any streaming source you choose is likely to only address a minority share of your total readership. And if you decide to pick YouTube for this, well, good luck: It is not built for listening on mobile without the upgrade to Red, does not integrate with your local streaming tools like playlisting, saving, or liking, and pays artists a fraction even of what Spotify’s free tier does. You should share songs in the format that serves your readers, the artists, and your own goals.
This solution goes beyond creating a universally personalized experience for listeners. Artists are largely compensated by streaming services through monetized plays, but if most of the links you encounter come from listeners with different streaming services from you, that means most of the traffic driven to a band’s dominant revenue avenue is bouncing off a brick wall. We could maximize the number of monetized plays for artists by making sure listeners see the right player all the time. This goes even deeper for labels as they respond to the growth of streaming by forming marketing strategies around it: They need to ensure they reach the greatest number of streaming plays to maximize returns. That is what a universal player could do.
We realized this was more than just a way to make Cymbal’s vibrant community more accessible — in fact, Cymbal’s app would be our first successful case study, a feature of a wider vision to restructure the way we listen everywhere on the internet and beyond.
Quietly, Cymbal integrated the Apple Music API and began to match songs. By the time of this writing, over 500,000 songs have been matched, with new matches created every time a user shares a song. Then, in April of this year, we brought our matched songs to their second medium, the web, in the form of a Universal Player, and built in crowdsourced matching for SoundCloud songs to expand our offering. It is a very simple tool: The first time you encounter it, it asks you which streaming provider you subscribe to and stores your preference so that every time you see the player, anywhere on the internet, it will always show you the song you have encountered on the player you pay for.
We leveraged Cymbal’s existing network to begin to scale the tool. First we built it into our social sharing, where our users post it to Twitter and Facebook hundreds of times a day. Next we integrated it into our New Song Alert emails, where we notify our users of new music from artists they have shared in the past. This already reaches tens-of-thousands of subscribers every week, sometimes that many in a day. Then we built an iMessage app with the same thrust: Want to send a friend a song? Search for it easily and make sure she can listen to it without knowing anything about what her streaming choice is.
Just a few weeks ago we even built a search engine for songs and pushed everything off our homepage to make space for it. Give it a spin: You can search any song, hit share, and drop it anyplace on the internet. We are already working to power streaming on subreddits and have our eyes set on anywhere music is shared: Social networks, blogs, chats, and beyond.
Cymbal’s goal is clear to us now. It is our job to make music social, and this can only be done if we all speak the same musical language. We are going to power all of the listening on the internet.
There may not be an opportunity to win the streaming wars; that fight seems destined for intractable stalemate between trailblazers and institutional giants. There is a huge opportunity, however, to make the entire internet listenable by becoming its musical translator; to connect listeners of all stripes to the experiences and communities that matter to them; to collect and understand streaming music data, regardless of its placement or provider of origin. Music must be playable to anyone who pays for it, but from any vantage point in streaming, it isn’t. Until now.
Again and again, I think of Valerie Miller. She was part of a team that saw an obvious barrier to communication, and her work was so foundational that it is taken for granted, even as texting became ubiquitous. Quietly, she helped make the world social at a scale that is hard to surpass, building something so fundamental its difficult to imagine a world without it. She built a bridge. Until the bridge is built, there can be no journey.
