Adding a feature: Spotify listening parties
Ironhack UX/UI Bootcamp — Week 4

Tools: Survey, Jobs to be done (JTBD), Value Proposition Canvas, Atomic Design, User Flow, Low-fidelity, High-fidelity, Prototyping, User Testing.
Duration: 4 days.
Technologies: Pen & paper, Sketch, InVision.
Team: Just me!
Introduction
This week at Ironhack Lisbon I had to add a feature to a pre-existing app. My colleague Vivian Reis asked me to add a “Listening party” feature on Spotify.
I started by trying to understand what led her to ask for this feature. Let’s get a bit more of context:
- Vivian is a Brazilian DJ who recently moved to Portugal. She often DJ’s at parties in Lisbon;
- Many of her friends are still back in Brazil, and she wishes they could attend parties where she DJ’s;
Therefore, the creation of a listening party feature would be ideal for her to share this experience with friends all over the world.
Research
I wanted to understand if people were familiar with the concept of a listening party and assess if this is something they’d be interested in doing on Spotify. So I created a survey and got a total of 72 valid answers.
From there, I was able to collect the follow insights:
- 84.7% of people have never participated in listening parties;
- 67.2% weren’t familiar with the concept of listening parties;
So just to clear things up: what exactly is a listening party?
“Listening parties are loosely directed gatherings aimed at discovering new music and discussing about it. Listening parties could also be a special event held to listen to a new release with like-minded friends or would-be fans of the artist.” — (Ehow)
Furthermore, the main reason for people to join listening parties would be connection, empowerment and discovery of new music.
- 80.6% said they listened to music primarily on their phones;
- 54.5% said that the two main reasons they’ve participated in listening parties is to discover new music, album releases and to hang out with (new) friends;
Jobs to be done (JTBD)
During this project we were introduced to a new User Research methodology — the Jobs to be Done. The basic idea is that every app is “hired” to do a specific job for a costumer.
Keeping this in mind, I defined 3 JTBDs that this feature would respond to:
- When I create a listening party I want to be able to share it with my friends abroad so that I can share the experience with them.
- When I listen to a new album release I want to invite my friends to join the listening party so that we can discuss it together.
- When my friends create a listening party I want to be able to join so that we can share a memorable experience.
Value proposition canvas
As a complementary method, I created a value proposition canvas to better understand my costumer. This tool was essential to understand the gain creators and pain relievers that the service would bring to the table.

Low-fidelity screens
My next step was creating a lo-fidelity screens of what the user flow could look like, and then testing with potential users. From here, I was able to get valuable insights on how to improve the concept before moving to hi-fi.
The main insights that I got were the following:
- Some people that I tested with were unsure how to reach the listening party feature, and would immediately try to “search” for it.
- Some people would successfully create the “listening party”, but would forget to share it afterwards.

High-fidelity screens
When I moved to the high-fidelity screens, I fixed these issues by doing the following:
- Asking users to create a listening party from a pre-existing playlist.
- Encouraging people to share immediately after creating the party, by adding a CTA (Call-to-Action) pop-up screen.
Atomic Design
A very important part of moving on to the hi-fidelity screens was that, for the first time, we decided to dive into a new methodology for creating design systems: Atomic Design. As the name suggests, it comes from an analogy with chemistry that defines objects in “building blocks” (such as atoms, molecules, organisms, templates and pages).
This was especially relevant because I had to break down the different components of Spotify’s system, analyze them and try to reproduce them in my hi-fidelity screens.
Prototype
Check out the prototype I created with Invision and tested with users. If you have the time, please test and let me know what your thoughts were!
Happy path
I’ve also included the video of the prototype with the “happy path”, i.e, the flow the user would take to get from point A to point B with no error. Check it out if you’re interested :)
Presentation
On Friday, I presented my final presentation to my colleagues. I was overall happy with the results!
Main learnings:
- For this project we worked with the Atomic Design methodology, which I found very interesting and useful. We also learned how to create “symbols” and “text styles” on Sketch, and work with libraries!
- I had to do thorough research on how the Spotify design system works. It was very useful to discover resources on Spotify Design.
- Regarding next steps, I would like to design more screens for this project. I thought of a second flow where the user would look directly into search and could access public listening parties from there.
- It’d also be interesting to integrate the Spotify messaging feature that Abia Karim designed.
I’m very satisfied with how this first UI week went, and am ready to keep going!
