Spotify Fit — Add A Feature
As ever evolving people, its hard to find something that changes with us, but it doesn’t have to be.
“Hot then you’re cold, yes then you’re no, in then you’re out, up then you’re down.”
In Katy Perry’s hit song “Hot N Cold” she expresses her frustration that we, as humans, change our minds like we change clothes. Clothes and music are similar in that they are both forms of self-expression. And as ever-evolving people, its hard to find something that changes with us, but it doesn’t have to be.
This week, at Ironhack Miami’s Design Bootcamp, we were tasked with choosing a popular application and incorporating a new feature. I chose Spotify because I am a loyal daily user and I was curious about where I could leave my mark on it.
To add a feature, I followed the design thinking process. These phases include: Discover, Define, Ideate, Develop and Deliver. After which, we will get into my final thoughts.
DISCOVER
The first step I took was doing secondary industry research in order to understand the client and whats going on in their market. I gained a few key insights: For the first time ever streaming revenue accounted for 56.1% of global recorded music revenue. A consequence of this was the decline of -5.3% in physical revenue. Additionally, Spotify is reigning supreme in the global music subscription market, with 130 million subscribers. Meanwhile, Apple Music coming in second place at 72 million. (Source: Statista)
This is where we step into business analysis, I began with the competitive feature comparison chart which gave me an understanding of our strengths and weaknesses compared to our competitors. The three highlighted areas all fell under the theme of increasing entertainment value.
It became apparent that there are not many popular free platorms for uploading music, which is why a lot of people use Youtube Music or Soundcloud in conjunction with Spotify. Additionally, other competitors have personalized playlists, but Spotify currently does it best. This is a big reason why users love Spotify. Lastly, I found that many people jumped to Apple Music and Youtube Music seeking video content, which Spotify is lacking in.
With the same competitors, I placed them in a market positioning chart in order to have a clear visual of where existing products stand in the market, so that we can decide where we would like to be. In my mind, it is comparable to a game of chess.
I began by identifying factors to entertainment apps that people consider, before joining, and came up with the axes below. In the chart on the left, Spotify is very popular already, it would just need to increase its entertainment value. For the chart on the right, Spotify would just need to provide more personalization options to the user. Thinking about this revealed two blue oceans, which are areas of uncontested opportunity, this show us where we want to be. As you can see, we are pretty close, so all we need is that feature to put us in that spot. Being in that spot, makes Spotify standout as an innovator.
Considering the secondary research, competitive feature comparison chart, and market positioning chart, I felt that I was ready to fill out the lean UX canvas. A tool used to visualize and record the problem areas of both sides, the business and user.
In this tool I came to the assumption that the main problem the business currently has and that the users feel is concerning playlists.
To validate or invalidate this assumption, I moved onto user research. That began with collecting quantitative data which took form of a survey completed by 43 people. The goal was to discover the problem that would be most impactful for the most amount of people. With doing this survey found out who my audience is.
I found out that I needed to focus on solving a problem that had to do with music.
I would focus on serving the majority, and those people listens to music every day.
I found out that I would need to make my feature on mobile, since, again, the majority of people use their phone to listen to music every day.
Now I knew what type of user I would most likely be interviewing and getting that qualitative information that is revealing. Interviewed five people, and here are a few of the most insightful quotes I was able to record.
I love using playlists, but I rarely do it because I don’t want to sit down, and take the time to make them. — Victoria
One user, Victoria, said she loves playlists but rarely uses them because theyre time consuming to create
I don’t make playlists because its annoying, I just want to just enjoy the music from what I already have. — Ariel
Another user, Ariel, said that he finds it annoying to make playlists because it takes time away from enjoying the music he already has saved.
I select a “Jazz playlist (…) and the mood is not consistent throughout the playlist. — @beerbistro on Spotify
The last quotation is actually taken from Spotify’s suggestion box page, where one user detailed how when he chooses a playlist he is frustrated because the mood is not consistent throughout.
It was hard to not ideate right then and there with all this information, but in order to best solve the problem, I had to define it clearly.
DEFINE
I used an affinity map to lay out all the data, qualitative and quantitative, that I received from the Discover phase and re-organized it to discover patterns and trends. The most interesting themes that I discovered had to do with what people were doing when they listened, what playlists they use, and what pains they currently have.
At this point, I had a value proposition for the user. I made this chart in order to depict my understanding of what the average user would need and value.This is done by looking at the data, and identifying a few things: The “Customer Jobs” which makes us consider why the user would want to use our product, the “Pains” which highlight their current frustrations with the experience, and the “Gains” ensuring we add value to their life.
In order to start improving the experience, we had to see where we were lacking. That’s where the as is scenario map came in, to highlight those dips in the experience. This one follows a person about to work out, but ultimately giving up because they are frustrated, they cannot access a consistent queue of music. Without their music they are unmotivated. You can see that their mood decline as the phases continue. This experience leaves the user feeling worse off than they were before they came into contact with Spotify.
Following that same journey, I added more detail in order to create a user journey map. The detail is invaluable in providing context, and highlighting areas of opportunity. I found the main problems to be with accomplishing functional and emotional tasks.
With the three areas highlighted above, I created a problem statement from each of them, and then reworded them into separate how might we statements:
- How might we allow our users to quickly make custom playlists?
- How might we give users freedom to experiment with their music?
- How might we allow users to define the music genre based off their mood?
Doing this turns these problems into questions that invite ideation.
IDEATION
It was soon time to brainstorm any and all possible solutions to our users problems. The goal is to reveal areas for design, and to do that I thought up about 19 ideas.
DEVELOP
Now how are we supposed to filter them out? Well, with both the business and the user in mind, we think about what solutions will best fit our constraints. A.K.A. feature prioritization! We want to make sure that the features are highly impactful and add value to the customer experience (the y axis). All the while being manageable for our business to accomplish (the x axis). Each quadrant has a name, but we want to focus on the top two, the left being the must haves, and the right being should haves.
The goals for the new feature are defined now, but what will it actually do? On the feature side of the value proposition canvas, we ensure product market fit; having strong market demand. Each section corresponds with the customer side. So, again, the “Products & Services” goes to fulfill the “Customer Job”, the “Pain Relievers” remedy the customer “Pains”, and the “Gain Creators” ensure we add value to their “Gains”.
To synthesize both the user and feature sides of the value proposition canvas we turn them into job stories. This job story being why people seek out that service, and the feature job story allowing us to imagine what the users will gain from using our product.
With data, ideas, and exercises we make it to our minimum viable product statement. This tells us what our feature, at the very least, should be able to accomplish. For this project, it is:
A feature that lets users to quickly make custom playlist that they can control by mood and genre allowing for creative freedom and endless possibilities.
Before this product can solve anyone’s problem, we need to solve the problem of how to create it. A quick flow chart allows us to map out what we would need to build in order for it to be successful. We do this because it is quick, cheap, and meant for iteration.
DELIVER
Once we can follow the flow logically we can move onto the deliverables for this project. Using the flow chart for reference, I created a Lofi paper prototype.
Amazingly so, we can bring paper to life use maze. I had five people do a usability test, there to validate or invalidate my drawn experience flow in any way possible. These are the 3 screens with the most error.
This screen had the most misclicks at a rate of 60%, and hearing my interviewees thoughts out loud, they said they all wanted to just see what this paper Prototype was capable of doing. I explained that it was capable of doing much more than I’d created, but everyone I interviewed said this was a lot of fun so I learned that more people want to do testing than I thought.
This is the second screen with the most misclicks, and when I asked my testers about it, many said its a reflex to just hit dismiss on notifications, even when they are interested in the feature. I thought about moving the button, but Ultimately decided against it because I wanted to be consistent with Spotify is currently layout.
The last page I had the most problems, with was this one where the instructions are to play a song. I should have specified to play a song on shuffle, because I hadn’t considered that you could just click on the song in the playlist. This made me realize I need to be more direct in my instructions and consider other paths.
Using these learnings, I went on to build my mid fis, while implementing the lessons I learned from lowfis.
The playlist screen again had one of the highest misclick rates because there are multiple ways to play a song. In this version i had the songs and play button as active spots. The reason for the misclicks more to the left is because one user said that she mistook the circles for songs, and wanted to see if it was a trick.
Of course, design is supposed to be intentional, and I knew that I messed up here by having the user second guess themselves, but moving to the hifi I assumed it would be solved with more detail.
The other highest miss click rate, was the walkthrough screen. This is because I did not add a darkening overlay for the bottom portion of the screen. Rather than focusing on the next button, especially with it being gray at the moment, it was easy for people to get confused and click on where the border was.
The last issue pointed out to me, was on the intro popup where users were confused where to click. I can see how testing with mid fis are helpful, because color really does direct attention. I again assumed that once I moved to hifi, this problem would be solved.
FINALLY — Introducing Spotify Fit.
A playlist generator that allows users to create custom playlists which they control by setting the mood and genre providing creative freedom. Like I mentioned at the beginning , we change our minds like we change outfits, so why not have your music reflect your fit.
The walkthrough follows the path of a daily active user coming to the app after they have updated it. Please click on the link in order to see it full screen and in action!
https://xd.adobe.com/view/f0ef911c-d0d4-4678-972b-12e0163af572-a5bc/?fullscreen
Within this demonstration, you will notice the atomic design elements. The elements are each individual design component. We compile these into one organized layout so that we can make the web developer’s life a lot easier.
With a new feature introduced to the market, how will we measure results?Done with the help of Success and Failure Metrics, it outlines what each result would look like for this project.
It would be successful when theres more engagement and more sharing of playlists, with less time making playlists. People will feel more fulfilled being able to creatively express themselves. Failure would look like the exact opposite. Additionally, people would feel stuck not being able to have the liberty to express themselves as they wish.
After all was said and done, I considered what I would do differently if I had more time, and what I would do next. That being that I would want to collaborate with developers early on, get concept validation from users after ideating, and getting the high fidelity prototype tested.
Final Thoughts
This week was chaotic, to say the least. Despite the ups and downs of my process, from adjusting to new tools and methods, I am extremely happy with the way this came out. I mean, from a 4-day sprint, this is amazing!
Though, there were more times than I’d care to admit that I just wanted to melt. Letting go of perfectionistic tendencies is tough, but the reward of contentment is worth it. Sometimes, you just gotta say: “it is what it is”, and let it go. Perfection is a facade and has no room for the complexities we humans possess.
Additionally, I would have never thought I would be doing anything like this. Or better yet, even be capable of doing something close to this. In short, this week was full of learnings and suprises about the designing process, as well as my own capabilities.
Week 4, done.
