For Ü
A personalized recommendation strategy that makes Amazon Music part of everyday life
Challenge
Design a Customer Experience (CX) strategy to improve retention across Amazon Music’s tiered offerings.
Problem Statement
How might we make users feel that Amazon Music really understand them?
Duration
4 weeks
Team
Allison Chen
Lanting Ko
Skills
User Research
Customer Lifecycle Analysis
Product Thinking
Prototyping
Tools
Figma
Google Survey
Introduction
Music that made for you
For Ü is a personalized recommendation strategy designed to enhance user retention on Amazon Music by addressing the emotional and behavioral needs of casual users. I collaborated on both the design and research phases, contributing to survey design and analysis, user interviews, usability testing, and prototyping. Throughout the project, I applied product thinking to ensure our solutions aligned with user needs as well as Amazon Music’s business goals. Our goal was to create a music discovery experience that feels timely, emotionally relevant, and effortlessly integrated into users’ daily lives.
Problem
Personalized content is the key factor of retention
Our research revealed that users highly value personalized recommendations. 29% of casual users reported disengagement due to poor recommendations, and 75% of lapsed users said they rarely use the platform because competitors offer a more personalized experience. When users encountered repetitive or generic playlists, they felt emotionally disconnected from the platform and gradually stopped using it.
Goal
Create an experience that feels relevant, personal, and engaging
We saw an opportunity to improve retention by delivering personalized music recommendations that align with each user’s emotional state, habits, and daily context.
Users
Casual listeners play a key role in improving long-term retention.
Casual users present the highest potential for long-term retention through improved design. While loyal users are already engaged and churned users require marketing efforts, casual users are still active and more easily influenced by the quality of their experience.
Persona
Understanding Two Listening Behaviors: Active vs. Passive
Music listeners interact with streaming platforms in different ways depending on their personal habits, routines, and intentions. Through user interviews and observation, we identified two dominant listening behaviors: active engagement and passive consumption.
Research
What we learned from users
To better understand the needs and frustrations of casual users, we conducted a survey with 113 participants and in-depth interviews with 5 users. Based on our findings, we identified seven key insights that shaped our design direction.
Users expect music that matches how they want to feel.
Music is often tied to memory, social connection, and personal expression.
Feeling understood by the platform builds emotional trust and keeps users engaged.
Personalized recommendations encourage users to explore beyond their usual habits.
Lyrics transform sound into meaning and deepen users’ emotional connection to the music.
Hypothesis
If Amazon music Personalizes, Users will Stay
If Amazon Music delivers a highly personalized experience, user retention will increase. This is because personalization builds emotional trust and a stronger connection to the platform.
Problem Statement
How might we make users feel that Amazon Music really understands them?
Design Decision
Make users feel understood
From our survey, we learned that playlists and libraries are key retention drivers:
50% of lapsed users would return if they had better access to playlists
53% of casual users use the platform mainly for this reason
We also found that users feel understood when music recommendations align with their:
Preferences (what they like)
Current emotions (how they feel)
Daily routines (when and where they listen)
Solution
Create a dual-track personalization strategy
System-driven Personalization
Preferences
Customer behavior across the Amazon ecosystem
Curated lyrics based on listening history
Emotions
Playlists that adapt to time, location, and weather
User-driven Personalization
Emotions
Turn mood into music with an image
Daily Routine
Customize daily listening based on routine
Daily Routines: User-driven Personalization
Customize daily listening based on routine
For Ü automatically generates routines based on user location and behavior, then recommends playlists accordingly. Users can also edit their routines as needed.
Emotions: User-driven Personalization
Turn mood into music with an image
Users can create a playlist by uploading images that best describe their mood.
Preferences: System-driven Personalization
Personal recommendation from the Amazon ecosystem
For Ü curates playlists based on users' Amazon service history and users can opt out in Settings.
Preferences: System-driven Personalization
Lyrics curated from the listening history
For Ü curates playlists based on users' Amazon service history and users can opt out in Settings.
Emotions: System-driven Personalization
Playlists that adapt to current time, location, and weather
For Ü adapts to users' location, time, and weather to provide playlists accordingly.
Differentiation Strategy
Make Amazon Music Stand out from Competitors
Results
Users rated their likelihood of returning to the app as 7.8/10 for tomorrow and 7.0/10 for the next month.
We interviewed 5 users to evaluate the retention potential of our design.
“I will go back to the app to see recent soundtrack from the show I recently watch.”
“It’s cool to create playlists with AI. I would like to try different photos.”
“I like the design that I can see and edit my routine because I usually listen to music based on the scenario.”
“Sometimes, images say what I can’t put into words.”
© 2025 by Lanting Ko. All rights reserved.
© 2025 by Lanting Ko. All rights reserved.
© 2025 by Lanting Ko. All rights reserved.
For Ü
A personalized recommendation strategy that makes Amazon Music part of everyday life
Challenge
Design a Customer Experience (CX) strategy to improve retention across Amazon Music’s tiered offerings.
Problem Statement
How might we make users feel that Amazon Music really understand them?
Duration
4 weeks
Team
Allison Chen
Lanting Ko
Skills
User Research
Customer Lifecycle Analysis
Product Thinking
Prototyping
Tools
Figma
Google Survey
Challenge
Design a Customer Experience (CX) strategy to improve retention across Amazon Music’s tiered offerings.
Problem Statement
How might we make users feel that Amazon Music really understand them?
Duration
4 weeks
Team
Allison Chen
Lanting Ko
Skills
User Research
Customer Lifecycle Analysis
Product Thinking
Prototyping
Tools
Figma
Google Survey
Introduction
Music that made for you
For Ü is a personalized recommendation strategy designed to enhance user retention on Amazon Music by addressing the emotional and behavioral needs of casual users. I collaborated on both the design and research phases, contributing to survey design and analysis, user interviews, usability testing, and prototyping. Throughout the project, I applied product thinking to ensure our solutions aligned with user needs as well as Amazon Music’s business goals. Our goal was to create a music discovery experience that feels timely, emotionally relevant, and effortlessly integrated into users’ daily lives.
Problem
Personalized content is the key factor of retention
Our research revealed that users highly value personalized recommendations. 29% of casual users reported disengagement due to poor recommendations, and 75% of lapsed users said they rarely use the platform because competitors offer a more personalized experience. When users encountered repetitive or generic playlists, they felt emotionally disconnected from the platform and gradually stopped using it.
Goal
Create an experience that feels relevant, personal, and engaging
We saw an opportunity to improve retention by delivering personalized music recommendations that align with each user’s emotional state, habits, and daily context.
Users
Casual listeners play a key role in improving long-term retention.
Casual users present the highest potential for long-term retention through improved design. While loyal users are already engaged and churned users require marketing efforts, casual users are still active and more easily influenced by the quality of their experience.
Research
What we learned from users
To better understand the needs and frustrations of casual users, we conducted a survey with 113 participants and in-depth interviews with 5 users. Based on our findings, we identified seven key insights that shaped our design direction.
Users expect music that matches how they want to feel.
Music is often tied to memory, social connection, and personal expression.
Feeling understood by the platform builds emotional trust and keeps users engaged.
4.Personalized recommendations encourage users to explore beyond their usual habits.
Lyrics transform sound into meaning and deepen users’ emotional connection to the music.
Hypothesis
If Amazon music Personalizes, Users will Stay
If Amazon Music delivers a highly personalized experience, user retention will increase. This is because personalization builds emotional trust and a stronger connection to the platform.
Problem Statement
How might we make users feel that Amazon Music really understands them?
Design Decision
Make users feel understood
From our survey, we learned that playlists and libraries are key retention drivers:
50% of lapsed users would return if they had better access to playlists
53% of casual users use the platform mainly for this reason
We also found that users feel understood when music recommendations align with their:
Preferences (what they like)
Current emotions (how they feel)
Daily routines (when and where they listen)
Solution
Create a dual-track personalization strategy
System-driven Personalization
Preferences
Customer behavior across the Amazon ecosystem
Curated lyrics based on listening history
Emotions
Playlists that adapt to time, location, and weather
User-driven Personalization
Emotions
Turn mood into music with an image
Daily Routine
Customize daily listening based on routine
Daily Routines: User-driven Personalization
Customize daily listening based on routine
For Ü automatically generates routines based on user location and behavior, then recommends playlists accordingly. Users can also edit their routines as needed.
Emotions: User-driven Personalization
Turn mood into music with an image
Users can create a playlist by uploading images that best describe their mood.
Preferences: System-driven Personalization
Personal recommendation from the Amazon ecosystem
For Ü curates playlists based on users' Amazon service history and users can opt out in Settings.
Preferences: System-driven Personalization
Lyrics curated from the listening history
For Ü curates playlists based on users' Amazon service history and users can opt out in Settings.
Emotions: System-driven Personalization
Playlists that adapt to current time, location, and weather
For Ü adapts to users' location, time, and weather to provide playlists accordingly.
Differentiation Strategy
Make Amazon Music Stand out from Competitors
Results
Users rated their likelihood of returning to the app as 7.8/10 for tomorrow and 7.0/10 for the next month.
We interviewed 5 users to evaluate the retention potential of our design.
“I will go back to the app to see recent soundtrack from the show I recently watch.”
“It’s cool to create playlists with AI. I would like to try different photos.”
“I like the design that I can see and edit my routine because I usually listen to music based on the scenario.”
“Sometimes, images say what I can’t put into words.”
Persona
Understanding Two Listening Behaviors: Active vs. Passive
Music listeners interact with streaming platforms in different ways depending on their personal habits, routines, and intentions. Through user interviews and observation, we identified two dominant listening behaviors: active engagement and passive consumption.