UX Designer + Researcher
Music Discovery.png

Tastemakers

 Tastemakers

A UX case study in understanding music discovery and listening practices.

Summer 2019

Music Discovery.png

Overview

As part of Springboard’s UX Design Course, I conducted user research on music discovery and listening practices. Based on this research, I discovered that several users lack trust in music recommendations on major streaming platforms. This became the driving motivation to design a different kind of platform for music discovery — one that emphasizes human recommendations rather than algorithmic ones. Tastemakers is a speculative app that allows users to discover music recommendations from their friends. It also allows users to browse and share music from different platforms. In this case study, I share the research and design process that led to Tastemakers along with the things I learned about UX along the way!

Background

Millions of people across the world use streaming platforms to listen to music (Forbes). Streaming platforms like Spotify and Apple Music use recommendation algorithms to expose users to songs they are more likely to engage with (Harvard Business School). These algorithms learn about user’s preferences based on their listening histories -- which means that they recommend songs that users are either already familiar with or sound similar to what they’ve already listened to before. However, users have expressed dissatisfaction with these recommendation algorithms and the personalized playlists they generate (Reddit), which suggests the potential need for incorporating different modes of music discovery on these platforms.

User Research

I began by sending out a survey in order to get a broad understanding of the various way people discover and share new music. Surveys were sent to the mailing list for my local college radio station as well as friends and family. My goal was to get varied responses from “experts” and “non-experts” — in this case a combination of radio DJs, avid music listeners, and occasional music listeners.

Results from the survey

Results from the survey

survey 2.png

Based on these survey responses, I prepared a set of interview questions to further understand the specific processes and painpoints that users have with regards to music sharing and discovery. I conducted interviews with 5 survey respondents who were willing to chat for 30-45 minutes. 3 interviews were conducted in person while 2 were conducted over the phone.

“I think with respect to music and web based discovery, the human element can sometimes be lost in the shuffle. That idea of a person curating and delivering seems to be a second thought or add-on feature.” - Interviewee 1

“Sometimes I’ll listen to someone else’s playlist if I feel like my music is becoming stale.” - Interviewee 2

“I have a lot of different things in different places, and it can be a pain. It would be nice to have a centralized list of stuff, even if it links to other services. Cause I have a lot of things on Spotify, then there’s a YouTube ‘just jams’ playlist. And I can never remember if one song is on another.” - Interviewee 3

“I think [Apple Music] was tailored to people who knew what type of music they wanted, but I was exploring” - Interviewee 4

Affinity Mapping + Empathy Mapping

After conducting interviews, I use affinity mapping to gather insights and organize them into different themes. Affinity maps allowed me to uncover some of the more prominent insights, which were then incorporated into three distinct empathy maps:

  1. Play-it-safe guy: a less avid music listener who generally doesn’t do much exploring on their own

  2. Music nerd: an expert music curator who listens to music on several different platforms

  3. Social adventurer: an avid music listener who enjoys sharing music with their friends

Map 1.jpg
Map 2.jpg
Empathy Map 1.png
Empathy Map 3.png

Personas + HMWs

Based on the empathy maps, I developed three personas to establish the kinds of users I would be designing for. But before I got to designing, I generated three “How Might We” questions to think about the kind of experience I was looking to design.

Paul Persona 2.png
Dani Persona 2.png
Sam Persona 2.png

How might we generate more meaningful music recommendations?

 

How might we make it easier to listen to music across platforms?

 

How might we make music discovery more social?

Ideation

User Stories

process 1.jpg
process 2.jpg
user stories.png

Sketching

sketch 1.png
sketch 2.png

Information Architecture

Sitemap

site map.png

User Flows

Design

Paper Prototype

Usability Testing

Wireframes

wireframes 1.png
wireframes 2.png

Mockups

Sign Up.png
Song Page.png

Takeaways