Case study: Improving time spent listening to Spotify
Overview
Spotify is a digital music, podcast, and video service that gives you access to millions of songs and other content from creators all over the world. The User Experience and User Interface Design in Spotify is already great, the contents are outstanding and inspirational for music lovers, plus its excellent recommendation feature. However, there is always room for improvement, and that’s how this case study was born.
Tools
- Sticky notes & paper (for ideation)
- Figma (for prototyping)
- Maze (for testing)
Duration
4 October 2021–6 October 2021 (3 days)
Objective
Improve the time spent listening
Key Result
Increase the average amount of time per unique user ID spent listening to audio content per month from 490 minutes to 540 minutes.
Research/Analysis
Desk Research
As a music streaming service provider, Spotify is always updating its technology with current trends. One of them is Artificial Intelligence (AI)-based technology, which is useful for strengthening its dominance in the music streaming industry. AI itself is a science and technique of making intelligent machines, especially intelligent computer programs. This ties in with similar tasks that use computers to understand human intelligence, but AI doesn’t have to limit itself to biologically observed methods. As a generation of productive age who actively uses technology devices, the millennial generation has several characters when searching for and accessing music streaming services on the internet. This can be seen in research from Digital McKinsey, on The Beat of Progress: The Rise of Music Streaming in Asia. it is stated that music streaming service providers should provide local content so that it is more relevant for local users who access the service. Then personalization, service providers should provide music content based on user likes and preferences so that it is more personal in nature. This will make it more exclusive. and further makes it easier for users to access the music streaming service. Data from Spotify Indonesia Consumer Insight (2017), clearly explains how millennials consume Spotify content, namely before going to bed (37%), during study time (35%), while cleaning the house (29%), waking up (24%) (Spotify & Nielsen, 2017).
User Research
I am a user of Spotify and I am a satisfied user, what I like about Spotify is that the UX and UI are just that good. They always make improvements as well. But only looking from my perspective is not enough, to get to know more about the users, I went out and conducted user research. My primary method of user research was user interviews so that I can produce something that really needed by users and have a good impact on business.
I decided to do a short interview with 5 of my friends while we are in the coffee shop, they are Spotify users and always share something from Spotify to their Instagram stories. I was curious about what they thinking about Spotify. Are they have any difficulties while using Spotify? What something that they really need but they can’t find in Spotify? and then I found some insights, I took a note and put it in the sticky notes, I separated between pain point and needs, and this is what I’ve found:
How Might We
The idea that came up to my mind after doing desk research and user research was :
“How to make it easier for users to find or play a songs/playlists/podcasts”
Wireframe
I made a rough wireframe using paper to capture the ideas that popped up in my mind based on the findings that I got. These wireframes are used as the basis for making a prototype design along with the ScreenFlow.
Interface Design
Play song for me: play songs/playlists/podcasts based on mood or activity
Often when we do a certain activity or are feeling certain feelings we want to play a song or playlist or podcast but don’t know which one to play. This feature allows users to play music based on the mood they are feeling or the activities they are doing easily because Spotify will play songs that are very suitable for us in our current condition.
Search song by sound
Have you ever been somewhere and listened to a great song but you didn’t know the song title? Or do you ever remember a song and want to listen to it again but only remember the tune? This feature allows users to search for songs by sound, now the user experience in searching for songs will become easier and more enjoyable.
Research Validation and Feedback
Research Validation Result
I conducted concept testing to validate my design and get user feedback with Maze, a remote testing platform. Here is the link UX Challenge: Spotify Concept Testing in case you wanna be my respondent for the concept testing too!
I chose to conduct testing with Maze because it is really easy to use and I can gather feedback easily just by sharing the link, the testing report is also easy to read. Here are the results from testing that I conducted at Maze with 7 participants:
- Play song for me: play songs/playlists/podcasts based on mood or activity
Usability Score: 80
2. Search song by sound
Usability Score: 84
Feedbacks
During the testing, I’ve got feedbacks that are very useful for me to make improvement from the design that I’ve made, one of them is shown below:
“play something for me:
I think Spotify really needs this feature because in some cases my mood isn't in good condition but I have to do something, when I want to try to do it while listening to music with the intention of improving my mood, I’m even confused because there is no suitable music. One note, I’m not comfortable with the appearance when choosing a mood using emoji, because at first glance they all look the same (I was a bit confused before finding the mood I wanted to choose), maybe (IMHO) it would be better if it was compared to the appearance when choosing an activity
Search by sound:
It’s an interesting feature if it’s implemented on Spotify because so far when I’m outside and listen to music that I think is good, I just open Shazam and when I find it, I just play it on Spotify. It’s better “
I did improve my design, especially for the interface when users choose a mood, I add a short description for each emoji with the purpose user can just read it and not struggle to describe their mood.
Conclusion/Reflection
There are some constructive feedbacks and can be used as a provision to make improvements to the designs that I have made, these feedbacks are useful for correcting my assumptions as well as giving me another point of view. Even in conducting research, I only use desk research and user interview approach, which can be improved by using other research methods to generate ideas with strong validation. So far I enjoy the process of making Spotify case studies for this UX Challenge, it can improve my ability to design thinking approaches and create products that are really useful for users while increasing profits in terms of business.
Thank you for reading! If you have any feedback regarding this case study, I’d like to hear from you, feel free to put some comments. If you want to connect, say hello at nabellanissa@gmail.com or connect on LinkedIn.
