Did Spotify just destroy its own mobile app?
Spotify just released a totally new mobile app — sort of.
Spotify recently made its new “Stations” app available in the U.S. for the first time, whereas it had previously been available only to users in Australia starting in 2018. The new app lets users create “radio” stations that play music based on certain artists or genres that the user selects. Creating a station is as easy as choosing two or three artists and letting Spotify’s algorithms find similar music, a task which they perform quite admirably, and, it’s worth noting, have done for some time. Strictly in terms of functionality, the core radio aspect of this app is nothing new, but there’s much more to Spotify Stations than the fact that it includes radio.
Here’s the thing: Spotify has been exposing its users to new artists via artificial intelligence for years, and has been building features into its desktop and mobile apps to leverage its prowess in predicting what songs users will love. In the last few years Spotify has added a “Recommended Songs” section at the bottom of users’ playlists, highlighted “Radio” as one of the three main navigation defaults in its desktop application, and, most importantly in my opinion, made its software play music similar to your playlist once you’ve exhausted all the songs in your current playlist.
These algorithm-driven suggestion features are valuable to me in this order, high to low:
- Auto-playing songs after a playlist ends
This feature consistently impresses me with a fantastic ratio of songs that I really enjoy to songs that I want to skip. Makes sense given that it chooses songs based on the content of playlists that I have already curated myself. I’ve found myself more than a handful of times jamming out to an amazing song, one that I’ve never heard before, having become engrossed in something with friends and having forgotten entirely about DJ-ing. After realizing that the music playing is indeed not part of my playlist, I’ve internally thanked the product designers at Spotify that my phone A) kept playing music at all and B) kept playing music that fits what I had been playing before. When you realize that someone hundreds or thousands of miles away has read your mind and done exactly what you want, without your having to ask, it’s a great feeling. That’s what this feature does for me. Thanks Spotify product and UX people!
When one song really stands out to me I can leverage Spotify’s algorithms to my advantage — with the standalone radio feature, not so much.
2. Radio — in many forms
Radio is easy to use, and I enjoy the songs it plays based on specific artists, but I almost never find myself using standalone radio in my daily listening experience. One of the reasons I use Spotify, and have been using it for the last six years straight, is that I can play whatever song, album, or artist I want, whenever I want, whether I’m on a free plan or a paid one. As a power user, playing a radio station based on a specific song or group of songs is much more useful to me than a general radio feature based on artists or genres. When one song really stands out to me I can leverage Spotify’s algorithms to my advantage — with the standalone radio feature, not so much. The same applies to playlists. The vast majority of the time that I flex Spotify’s music-finding brawn is when I want to find more music similar to specific songs or playlists that I’ve already curated, not just to play certain artists and their sound-alikes. Yep that’s a new term. You heard it here first! Shhh...
3. Recommended songs — that list below the end of your playlists
Unfortunately, the recommended songs that appear below my playlists are much more loosely related to the songs in my playlist than say, that playlist’s radio station, and often tend towards more popular songs currently topping the charts or towards the songs in my playlist from the month before — songs that I’ve already listened to hundreds of times. Not super useful.
My listening experience is centered around the active curation of personal playlists — finding new songs with very specific sounds and putting them together in ways that are relevant to me, whether it’s by genre, sub-genre, mood, month, language, or another way of grouping songs that I want to listen to together. This need for personalized groupings is where radio and algorithms fail me; this is why I used Pandora briefly back in the day before realizing that a radio-like listening experience was not what I wanted.
Spotify, on the other hand, built a stellar listening experience around curating and playing specific music, on demand, anytime, anywhere, even on an ad-supported free version. Spotify did indeed have radio long ago (think 2010), but playlists and curation won out, rocketing the platform to the top of the music streaming market, buoyed by younger users who wouldn’t accept a radio-like listening experience in the age of two-day delivery, binge-watching, and information on demand. Why listen to the internet equivalent of a radio station when you can listen to the popular new rap album, the whole popular new rap album, and nothing but the popular new rap album?
Fast-forward five years and Spotify is absolutely crushing the curation and on-demand music game.
Fast-forward five years to 2015 and Spotify is absolutely crushing the curation and on-demand music game. More people around the world are using streaming services than ever before to consume music, movies, TV, and video games, which means tons of potential new users, but at the same time means more competition. The release of Apple Music begins to threaten Spotify’s on-demand dominance, and Pandora, ever the king of “radio” starts to offer on-demand listening features as well. Having a amassed a huge number of free-tier users, Spotify is aggressively trying to convert free-tier users to paid users by offering discounts to students and families, bombarding free-tier users with ads for Spotify premium, changing its free tier UX, and setting different prices for various regions to maximize revenue globally.
Today, in mid-2019, Spotify has in some ways maxed out. Engineers and data scientists still tweak and A/B test endlessly to eke out slightly higher conversion rates of free-tier users to paid plans, not to mention sign-ups and long-term engagement, but at this point the platform is extremely usable and offers an outstanding, if not unparalleled, user experience for on-demand listening. There are major gains to be made in terms of making the platform social and podcast-friendly, but these improvements still fall under the umbrella of on-demand listening. Though podcast listeners are admittedly a distinct and increasingly important segment that Spotify must chase, there is no reason that podcast listeners would need an entirely different listening experience from the existing one. To the contrary, I would argue that in most cases, because podcasts are series, podcast-listeners are served well by the browse-and-curate model that the platform already supports beautifully.
Spotify had to break its own mould and create a totally separate listening experience that would appeal to older users.
This brings us to the reason for Spotify’s new “Stations” app — the playlist curation listening experience limits the number of users Spotify can delight at one time. To reach new users and drive revenue growth, Spotify had to break its own mould and create a totally separate listening experience that would appeal to older users, the less-technically-inclined, and those users who stuck with Pandora all along because they simply didn’t care enough to curate their own playlists. In 2018, only 34% of Pandora listeners were under 35 years old, versus 55% for Spotify listeners. This more-passive radio listening experience, which Spotify has been supplementing with pre-curated playlists, new language and cultural hubs, and the algorithmic suggestion features that I mentioned earlier, was already available on the platform if you looked closely. This more-passive listening experience has not, however, been a central part of the product experience up to this point, and for good reason. The closer to Pandora Spotify’s product gets, the further from its strengths, its identity, and its core users’ needs it strays.
My mom just wants to put on some tunes and continue working. She’s never going to use a queue or actively curate playlists.
So how to bring into the fold those potential users who will really only benefit from a passive listening experience? It’s simple: Build a totally new experience for passive listeners. My mom just wants to put on some tunes and continue working. She’s never going to use a queue or actively curate playlists. The Stations app is perfect for this kind of user, essentially copying the radio functionality from the main Spotify app to the exclusion of all else. This results in an extremely clean and effortless listening experience that is focused around, well.. listening, rather than curating, queueing, or sharing.
The new app’s minimalist user interface supports only a small range of interactions and features. Gargantuan text for the names of radio stations makes it nearly frictionless for vision-impaired and/or older users to play the kind of music they love with just a swipe. Additionally, the app automatically plays music on being opened or switching stations. There’s no scrubbing control or full-featured player screen. Easy-peasy, dare I say it, lemon squeezy. Now this is a passive, effortless listening experience. Pandora who?
Spotify will continue to improve its core curation-based product experience, but in terms of winning the user war in a broader sense, Spotify’s Stations app is a fantastic play to capture casual listeners without degrading the experience of the core platform, while putting Spotify’s excellent preference recognition and prediction algorithms in the spotlight.
Whether those algorithms play second-fiddle to your library or not is up to you.. but at least have the decency to tell Mom.
