Spotify Concept: Design for Discovery

Spotify is the leading music streaming app available on almost any smart device. Like most users, I enjoy using Spotify to share, download and listen to a large variety of music.

Spotify has features for the purpose of discovering new music. However, it is hardly used because:

  1. It suggests irrelevant music that has no interest in the user.
  2. It suggests songs the user already knows about.
  3. It’s hard to share music effecitvely.

Spotify’s Discovery Page Struggles

Going into this case study, I wanted to observe the methods that Spotify used to discover and share new music. I can then compare Spotify’s methods to new methods that I set out to discover.

What Spotify uses:

Sharing, Accounts, Daily Mix, and Suggested Playlists

Understanding Spotify’s Flaws

User Research

My initial hypothesis was Spotify has poor music recommendations and sharing features, ultimately leading to inefficient music discovery. I interviewed 4 participants, each with a diverse background, to test my hypothesis.

To my suprise, I learned that most users did not even use Spotify’s implemented sharing feature but rather used their own incoventional methods to share music.

Sharing is an integral part to discovering new music and I decided to focus on the sharing aspect of Spotify to better the discovery aspect.

Brainstorming Session

In order to come up with solutions addressing the problem, I gathered a team of two of my friends to start a brainstorming session.

After an inspirational session, we narrowed down our problems into three main problem spaces:

  1. How might we better face-to-face interactions when sharing music?
  2. How might we create a better online music sharing community?
  3. How might make it easier to discover music you like?

From there, we picked the top 6 features [voted on by peers] that solved these problems in the most effective way.

Feature 1: Shared “music room” that people can connect to and see the music playing. The “aux” person is the leader of the room and can decide what songs are added to the queue but anyone can see what’s being played

Feature 2: Link to the current song playing on someone’s profile

Feature 3: An option that provides music from the selection: “People who listened to this song also liked…”

Feature 4: Suggest music based on information from other apps (i.e. Amazon, Facebook, Twitter, Games etc)

Feature 5: Be matched with friends who have similar songs on playlists

Feature 6: Artists create public playlists that people can access

Narrowing it down to Feature 1, 5 and 6, I conducted a guerrilla testing session which provided useful insight on which feature to pursue: Feature 1.

THE Feature 1

Almost everyone I know is always excited to find a song that they really like. The point of Feature 1 is to have people discover songs they enjoy in that easiest way possible. This feature essentially combines House Party and Spotify. In this feature, everyone on Spotify has the option to create their own virtual “lounge.” Depending on the settings, anyone or only friends can connect to your lounge. When a user joins the lounge, they can begin to listen to the current song playing in the lounge and see what songs are playing next (the queue). The user who created the lounge (aka the leader) has complete control over the songs being played in the lounge and the songs on the queue. However, all users that enter the lounge can have an impact on what’s being played by either liking a song already on the queue or adding a song to the queue from their own music library. The queue order is determined by the number of likes each song has (voted on by the people connected to the lounge), with the next song playing having the most number of likes, second song playing with the second most likes etc. To get an idea of what song is playing next, a user can hold down a song on the queue and it will play a short sample of the song only on that user’s device, without affecting the current song already playing on the lounge or the other users connected. Implementing Feature 1 limits the awkward face-to-face interactions of sharing music and people can express the type of music they enjoy listening to for others to listen. The expected overall outcome is to have users interact with each other, sharing and discovering songs wherever and whenever by just opening Spotify.

Laying out the framework:

User Flow

Go to Spotify Home screen → Click on “Connect to Lounge” → Arrive in the Virtual Lounge → Browse the queued songs and the current songs playing → Choose to like, save or add a song to the queue → Watch the queue continue to change as people vote on what they want to listen with new songs being added→ Save songs to your music library that you enjoy and discover new music!

Initial Designs

Going off the original Spotify design, I developed three entry points A, B,C to enter the Spotify Lounge. I also developed an early design for the Spotify Lounge.

The goal of the entry point was to allow for the user to have quick and easy access to available lounges without taking away from other features. The Lounge design was focused on keeping it simple: showing the song playing, what’s queded, and how the user can interact with the queue.

Spotify currently has social features limited to just following other users, seeing what type of music they are listening, playlists they created, etc. I wanted to implement an interactive social feature that allowed for interaction between users where everyone can have an impact on what’s being played. The purpose was for users to get a feel for what type of music their friends and followers are listening through the songs they have added to the queue and the likes that were distributed throughout the queue. This would hopefully give the user ideas for new music and allow for discover. Also, it limits any awkard face to face interaction that could occur when trying to share music.

Exploring User Flows

Flow 1 [User entering and interacting in a Lounge]

This user flow was designed around simplicity. It combines all the desired features into one screen, without overwhelming the user. Also, it is easy to follow and there are few steps, allowing for the user to access the lounge quickly.

Designing The Prototype

I converted my medium fidelities into high fidelities and used Invision to create the Spotify Lounge prototype. Immediately after running a handful user tests, I was left with a similar response: “How do I know who is connected to the Lounge?”

In order to address this problem, I went on to design an overlay feature that allowed the user to see who is currently connected to the lounge.

Making Improvements

The first user flow allowed for minimal user interaction and since sharing is an integral part of discovering music, I aimed to better the user sharing experience.

After redesigning the lounge, I attempted to stay with a simplistic feel and I added a button on the top left hand of the screen that when tapped would allow the user to see who is connected to the lounge on an overlay screen. From there, the user can tap a username and see the user’s profile. More interaction = more discovery!

Final Thoughts

Once I understood how big of a role sharing music has to do with discovering new music, it became easier for me to redirect my people problem into increasing user interaction while decreasing the negative aspects. This is where the idea of a lounge, where anyone can connect and share music, came to mind. The Spotify Lounge is far from reaching its full potential and I am always open to suggestions and would like any feedback as I plan to continue to work on this as a side project.