Back to top

Moodcast

Mobile app suggesting podcast based on user’s mood

Overview

This was a school project with the task to design a podcast app. Using online statistics as my starting point of concept development, along with discoveries from my own research, has inspired me to design a podcast app suggesting based on user's mood.

Role

Everything!

With who?

Tools

What did I do?

Research, ideation and design

Process

W1
Discovery
Desk research
User interviews
Behavioural archetype
Generating problem statement
W2
Research (Interviews)
Bodystorming
Generating ideas
W3
Wireframing
Lo-fi prototypes
W4
Testing
W5
Testing
High-fidelity prototype
Not interested in process? Click here to view final designs

Online research (Desk research)

To understand existing podcast trends, I have conducted research online, here were my findings that I found particularly interesting:

Only 90%

of Australia listen to podcasts, lower than the global average than 41%. (Sang et al, 2020)

Multitasking

More than half of podcast listeners consume podcasts while multitasking:"59% have listened while doing housework. 52% while driving. 51% while cooking. And 46% while going for a walk.” (The Edison Research, 2017)

The statistic on multitasking sparked my interest most, thus I decided to begin from there.

User research findings (Interviews)

To dive deeper into understanding the habits of podcast listeners, an interview was conducted with three interviewees who were occasional podcast listeners.

Multitasking

My interviewees multitasks while they listened to podcasts. Two of my interviewees listens to podcast while sleeping, while one listens while driving

Voice and content

they were two deciding factors into whether or not they would continue listening to a podcast.

App controls are difficult to access when multitasking

My interviewees multitasks while they listened to podcasts. Two of my interviewees listens to podcast while sleeping, while one listens while driving

Behavioural Archetype

2 behavioural archetypes were then developed based on my interviews:

Sleeping podcast listener
"I can't sleep without a podcast on"

Podcast listeners who listen to podcasts before they sleep or while they’re sleeping

Goal

Finding a comforting podcast for her to listen to before she falls asleep

Pain points
  • Manual controls: Making controls on phone is difficult.
It takes effort and is not the best experience when you’re falling asleep.
  • Podcast choice: Not being able to find a suitable podcast that fits their needs. Some users have a very specific taste when it comes to what helps them fall asleep.
Driving podcast listener
“Long drives to work are boring without something to listen to”

Podcast listeners who listens during their ong morning drive to work.

Goal
  • Quickly decide on something to listen to while he's driving
  • Queue 1-2 podcasts beforehand, so he won't have to manually choose a new one when the previous is finished.

Finding a comforting podcast for her to listen to before she falls asleep

Pain points
  • Manual controls such as volume control and switching podcasts can be difficult when he’s driving.

Defining the problem

After the interviews, I have gathered some key points and revisited my research afterwards.

A pattern I have found between these users was that they struggled with interacting with their podcast apps while conducting the task they were doing: Driving users can’t look at their phones while driving, while users falling asleep had issues with making interactions with their phones in the dark.

Opportunity for exploration #1

How Might We ease interaction between multitasking users and the app?

Another important point that stood out to me was when asked about what factors influenced their choice of the podcast, one commented on how the host’s voice always factors into whether or not they will listen to it or not.

Opportunity for exploration #2

How Might We increase confidence in the user's choice of podcasts, so interaction can be minimised between multitasking users and the app?

Not only should we enhance interaction for multitasking users, but we should make users feel more confident in their choice of the podcast, so interactions would be minimised.

Ideate

I have then took the pain points of users, and began ideating opportunities to solve these pain points:

Idea #1
Recommendations based on ‘mood’, not 'genre:
Many podcast apps in the market recommend users to podcast based on genre. My idea is creating an app that allows users to discover podcasts based on what the user is feeling. This would suit occasionally listeners (like my interviewees) that don’t have a podcaster they listen to on a regular basis, but rather just picks a podcast based on what they feel like.
Idea #2
Voice filtering
As one of my interviewees mentioned how voice is a factor of her podcast choice, we could design an app where users can filter podcast based on the podcaster’s voice (eg. high/low pitched, accented, gender)
Idea #3
Multitasking modes
Users can turn on a specific ‘mode’ (eg. sleep mode, driving mode) when they are multitasking. When using this mode, specific functions are enabled for a smoother podcast listening experience.

Research - Bodystorming

To narrow the scope of this project, I have decided to only focus on sleeping podcast users.

To have a better understanding of how it feels like to be one, I have conducted a body-storming session myself.
I conducted this exercise in my room at night. I put on my earphones, played a random podcast while falling asleep. I also tried interacting with my phone, adjusting controls, and changing podcasts.
The question I kept in mind the time I was doing this was:
Goal:

How can I design better experiences for these users so they can have a better night’s sleep, improving motivation and performance the next morning?

After the bodystorming exercise, I have outlined some thoughts and feelings, aiding me to create a solution later on:

Wireframing

I have then wireframed screens before putting into high fidelity prototype:

Final solution

My final solution is 'Moodcast', a podcast app that generates recommendations based on user's mood.

For multitasking listeners, they can enter specific modes that enhance their listening experience.

Because of time constraints, I have only developed a solution for users that listens to podcasts while falling asleep. Here is a full view of the final solution:

Fun onboarding screens

Designed to look fun while walking through users what the app does.

Pick and choose a mood

1. Users choose how they're feeling and a task they are doing

2. Recommendations are given based on this. (Eg. if the user is feeling ‘restless’, podcasts of meditation or ASMR will be recommended. If users ‘need a good laugh’, humour podcasts would be recommended)

3. Tasks dictate on what mode users will be taken to

Sleep mode: 
Helping users get a better sleep

  • If users click option ‘About to fall asleep' , they are taken to the ‘sleep mode’ homepage
  • Device turns to dark mode, enabled to protect listener’s eyes in the dark

Extensive filters tailored to different tastes

  • Filtering for voice, gender and content so listeners can more precisely narrow down what they’re looking for
  • A 30 second snippet preview of selected podcast, for more confidence in their choice of podcast

Soothing noises for a better sleep

On sleep mode, listeners can choose soothing background tunes for a better night sleep.

Minimal lock screen

  • A lock screen specifically designed for sleeping users
  • Minimal design with bigger buttons, without unnecessary to minimise cognitive strain with sleeping users
  • Minimising disturbance to user’s sleep while listening to their favourite podcasts

Seamless experience with Fitbit or Smart watch

1. Connecting electronic watches or Fitbit to app

2. When watch detects a declining heartbeat rate, volume starts to fade and turns off, saving battery and providing listeners with a better listening experience

Idea

Using data collected from listener’s's smart watches, the app could recommend better podcasts to other app users. For instance, if data shows users on the platform fall asleep quicker to a certain meditation podcast, then it will be more frequently recommended to other listeners facilitating better recommendations.

Takeaways and future considerations

This was a school project that spanned for a duration of 4 weeks where I was taught UX design thinking and put it to practical use.

If this was a real project. I would take time to do more research into podcast listeners, interviewing and observing their behaviour to identify needs and pain points.

Additionally, I would also conduct research into non podcast listeners, and explore on 'tasks' more, designing features for more multitasking listeners.

Technical feasibility would also be a consideration in a real project, such as how would voice be defined? and how would we generate recommendations based on moood?

More projects
Edifyd

Learning management System

Popsup

Contactless dine-in experience

Come and say hello!
About CharmaineCV