OVERVIEW
A CASE STUDY EXPLORING HOW SPOTIFY COULD UNLOCK THE POWER OF COMMUNITY.
YEAR
2019
CONTRIBUTION
UX // UI // RESEARCH
PROJECT TYPE
CASE STUDY
User research
Ideation
Explorations
Feedback
Solution
DESIGN PROCESS
I started with a broad research objective: understand core user needs and pain points in the Spotify music discovery experience. Why do people seek out new music? How people currently discover new music? What are the existing barriers to discovering new music?
USER RESEARCH
KEY INSIGHT 1: People have different needs for Spotify based on their relationship with music.
KEY INSIGHT 2: Spotify value props and pain points centered around Discovery, Organization, Usability, and Community.
KEY INSIGHT 3: Discovery and community needs were shared across the Functional Listener to Music Lover spectrum.
PEOPLE PROBLEM
Listening to and discovering music is a richer experience when shared with friends and family, but Spotify lacks the tools to unlock the power of community.
EXPLORATIONS
FRIENDS TAB
📍 Friends tab location in the menu bar felt natural
✨ Love the added context of Earlier
IDEATION
Ideation
Brainstorm
I recruited two participants for this brainstorming session. They are diverse from on another in gender, age, race, educational background, occupational expertise, and music streaming app usage. We started with a free-for-all brainstorm of solutions for the people problem, then grouped them into Community, In Real Life (IRL), Music Discovery, New Music Management, and Expression/Sharing.
Solution spaces
1. Friend aggregation
Aggregated list of friends, with categories that show who has similar taste in music
2. Playlists
Auto-curated playlist of top songs friends with similar music taste are listening to
Auto-curated playlist of songs recommended by friends
"Recommend this song" button to quickly recommend songs to friends
3. Metadata
Next to songs, artists, and albums, show which friends also like those things
Display individual music listening trends on each persons' profile (aka publicized Year in Review data)
Impact vs. feasibility
Aggregated list of friends + Auto-playlist of friends' top songs with similar taste: [Medium/high feasibility, high impact] aggregating friends and showing their top songs is fairly straightforward, though complicated by needing friend content to populate this section. High impact, as it takes visibility into friends' music taste from zero to one.
Auto-playlist of songs recommended by friends: [Medium/high feasibility, medium impact] song recs from friends are powerful, but less scaled
"Recommend this" button: [Medium/high feasibility, medium impact] recommending songs to friends is a powerful but less frequent need
Social context next to songs, artists, albums: [Medium feasibility, medium/low impact] adding social context to so many minute surfaces could quickly clutter the UI with non-actionable information
Music listening data on profile: [High feasibility, low impact] this would be really cool and fairly straightforward to implement since the profile and the data already exist – low impact because this information would be largely informational not actionable
Key Insight #3: Discovery and community needs were shared across the Functional Listener to Music Lover spectrum.
Explorations
Friend reccomendations
Friends tab
Friend-recommended song playlist
Combined friends + music tab
Separate friends + playlist pages
User Feedback
Three 45-min in-person qualitative interviews.
Questions
Do people find value in a friends tab?
What other data would they want to see in a friends tab?
Can people find their own and friend's profiles?
Do people find value in the updated profiles?
Are people able to successfully recommend songs to friends and locate recommended songs?
Do people find value in song recommendations from friends?
Scenarios
Find out what song a friend is listening to right now
Find out what friend current top songs are
Recommend a song to a friend
Locate the songs friends have recommended to you
Find out what your top songs are
Key Feedback
The feedback from crit mainly focused around the content of the friends tab. Right now, the friend tab feels stale and doesn't fundamentally change the social nature of the app. The idea for auto-generated friend playlists, for example elevates Spotify's AI rather than friend-to-friend connection. Digging deeper into the content issue, there's a tension between real-time and depth. Is the power of community stronger felt in knowing what your friends are listening to and when, or knowing more about your friends' overall listening habits and music tastes? I think the answer is that both are important for creating a vibrant social fabric within the Spotify mobile experience.
For my second pass, I focused on the core capabilities and content requirements needed to weave this social fabric.
“Example quote from someone about something. In reality, I have nothing profound to say, but I enjoy having my voice heard.”
Prototype 1: Profiles
Concerns about privacy
Like the anthem idea
Would like to see overlap of friends’ music taste
Prototype 2: Recommendations
Makes sense from 3 dot menu
Questions about where the recommendations go
Hesitation around how the recommendation will be received
Want more contextual info in friends list
Concerns around differing taste in music
Would like a combined section of all recommendations
Prototype 3: Friends Tab
Location in menu bar felt natural
Confusion around use of search bar
Desire to chat with people displayed as online
Learnings
Music is really personal.
Everyone has their own personal relationship to music that impacts their needs in a music-listening app.
Human connection breathes life into our experiences.
Even with an experience as personal as music, the magic is in sharing it with the people we love.
People love Spotify.
Every Spotify user I interviewed – of ranging levels of music obsession –expressed resounding love for the Spotify user experience. The care they put into building user-first experiences has paid off.
NEXT PROJECT