Reframing: Making a leap of faith into blue-skying
Another round of lo-fi, and then back to ideation
Hi there! This publication follows the MHCI Amazon Music team as they further explore the space of music by seeking opportunities in areas besides the “in real life” (IRL) experiences. We’ll be cataloguing our sprint-by-sprint process, as well as any insights we gain along the way.
For the past two weeks, the team focused on generating a second round of prototypes. However, before we got a chance to launch the prototypes to participants, we received some criticism from both our client and faculty mentors on the direction of the project. In an attempt to digest the advice, we reflected on our past trajectory, put an end to the current lo-fi explorations, and defined a new north star for the project.
For more information about our project, MHCI, and our team, check out our first blog post. ✨
Value proposition: algorithm cards vs. playlists vs. stations
Before diving into our new direction, let’s take a step back to where we left off. One major takeaway from the lo-fi testing on core flows was that users are not clear as to what the value of algorithm cards is in comparison to other formats. To provide a clearer value proposition for algorithm cards, we focused on defining key use cases where algorithm cards can triumph over playlists, albums or stations.
We first established that algorithm cards will be a better version of stations and that they will not replace playlists. Consequently, IRL use cases of algorithm cards will be our main focus because this is where playlists experience the most friction. In turn, the most valuable algorithm cards will emerge from users’ regular listening behavior.
Here is the new set of value propositions for algorithm cards. They will guide our process in the future on determining core features and prioritizing what to focus on.
- Users get to learn about their music identities.
- Algorithm cards play better and easier music for users in IRL scenarios.
- Algorithm cards reflect who users are right now - not who they were last week.
- Users easily can put on music that a group of different audiences can enjoy together.
Consolidating algorithm card features: card presentation and card editing
After consolidating the value proposition of algorithm cards, we explored how users can interact with VUI as a method to narrow down our scope. The VUI ideation landed on two features that could benefit from further exploration — card presentation and card editing. Iterating on the testing strategy, we aimed to make the test more effective by asking for feedback tailored to specific areas for improvement. Under this guideline, we then developed prototypes experimenting with a variety of methods for presenting information on the card. Concurrently, we also worked on a prototype to understand the most effective mechanism for users to provide feedback to better tune their algorithm.
Reorienting and reframing: where are we actually going?
Our prototypes, though well-crafted, were never able to make it to the testing phase. During our client meeting, we received the feedback that the features we are testing are too granular and that it is not the best use of our time to focus on usability at this stage. Similar sentiments were expressed by faulty mentors. The team took a hit, to say the least. After spending some time discussing the feedback and consolidating our thoughts, we reached the following conclusions:
Up until now, we have always worked with the underlying assumption that the end goal of our project is to deliver a fully fleshed out product — which means we would follow the conventional trajectory of iterating on the design with increasing fidelity. However, there exists a mismatch between our perception and the client’s expectation. Instead of creating a fully functioning product, our client hoped that we could showcase a final solution that would spark innovation rather than solve an existing problem. As a result, the team decided to put an end to our current GUI-based lo-fi exploration, and begin to ideate on novel ways that algorithm ownership can improve people’s music experiences.
Concluding algorithm card exploration: the system view
To cap and synthesize our lo-fi exploration, we created a system map, capturing core features of algorithm ownership and the backend data flow. Looking at the solution space from a system perspective allowed us to identify under-explored areas — for instance, there is an opportunity for us to expand on the current feedback mechanism to incorporate more modalities.
Blue-sky ideation: if not card, then what?
Although we will no longer limit ourselves to the format of a card, we aim to continue expanding on the possible applications of algorithm ownership. Situating the concept in the frame of “a day in the life,” we generated more ideas on how algorithm ownership can be used during IRL music listening scenarios. By coupling specific modes (emotion-based, activity-based, socially based, etc) with common daily routines such as waking up, commuting, work and winding down, we began to paint the image of a future where algorithm ownership is already achievable.
Some of the key ideations include:
1) Bookends Visualization / Planning Your Day or Seeing it Planned Out for You Musically
2) Environments being associated with unique sound fingerprint / Opportunity for unique interaction (i.e. physical perimeters)
3) Use of AR to ground music and expose it in reality / Sharing and social aspects
4) Tagging and Activities / Activity-Based Curation / Application is Broad and exciting
5) Customizing music with layers / Creating new interactions with music for users
Thanks for reading! Stay tuned for more updates from our team! 💖🎵👏
