Ariana Davies

SoundCloud users are frustrated by navigation and discovery of music flow

Users struggle to differentiate between albums, playlists, and songs. The streamline of finding where your saved music is and creating a playlist takes multiple steps and was noted an 8% success rate in tree testing. A complex interface forces the user click around before achieving their goals.

Consistency and Ease of Use shortcuts

Adding multiple avenues for creating a playlist increase task success rates by 24% and reduced time taken to complete said task by 82 seconds. Furthermore, adding icons to differentiate media types reported more user directness. Simplifying the navigational structure, increasing the contrast of elements, utilizing the SoundCloud orange to direct user attention, and implementing comment confirmation improved the success rates and confidence of users on SoundCloud.

Community shares music and experience…

Before conducting our heuristic evaluation, we dove deeper into the meaning of SoundCloud’s platform and how users view music today. In a report from BPI and Blackstar…

“From curated playlists on streaming services, to viral short-form videos, songs embedded in video games, TV shows and films, this generation is hearing music that is new to them in every conceivable way and open to discovery. Peer-group recommendations – through shared playlists and in-person gatherings – can be key to these respondents, forming communal soundtracks and social bonds.”

To make SoundCloud a top competitor for a new generation of listeners, we had to uncover potentially overlooked pain points to find what users may be having issues with. Using Nielson’s 10 Usability Heuristics as the basis for our evaluation, we determined the areas of friction for users with SoundCloud’s desktop website.

SoundCloud lacks where others succeed…

SoundCloud is uniquely positioned to advance in the market through its community-driven social features. By better understanding the competitors of SoundCloud, we will gather insights that may lead to a competitive advantage.

To validate our previous findings, gain more in-depth insights into user experience with SoundCloud, and test proposed design solutions in a short time period, we chose to conduct three tests:

Our two main focuses are:

92% of users failed to create a playlist

Users struggle to navigate SoundCloud, so we gathered 51 participants to conduct a tree test, focusing on core experiences of library navigation and playlist creation. You can find our full SoundCloud Tree Test Results here.

Majority of users found that homepage navigation was easy to miss

After learning through user mental models on navigation with our tree test, we had 6 participants conduct an unmonitored usability test to experience SoundCloud’s desktop website. This gave us attitudinal data on their reasoning behind decisions and surfaced additional friction points and positivities that were not visible through expert evaluation alone.

25 tasks were created for users to evaluate and share their opinions during their process, so we addressed major key findings with these associated tasks:

Success increased by 45% when adding a create playlist button into the playlist page

To validate our findings and test proposed designs, we opted to use a quantitative A/B test to compare user experiences between SoundCloud’s website and a tailored prototype based in our data from previous evaluations and tests.

As this was our biggest test yet, we opted for a larger group of participants using word-of-mouth and guerilla testing in places like coffee shops, university campus, and town squares. Test A gained 42 participants while Test B gained 43 participants, so we began synthesizing.

How might we create a consistent, flexible, and community-based music experience for SoundCloud users?

What could SoundCloud look like?

What Did I Learn and What Would I Change?

Looking back on this project, I learned a great deal of the importance of wording questions. If given the opportunity to do this again, I would’ve consolidated some of the questions within the qualitative usability test and used more specific language on sharing their perspective and initial reactions to SoundCloud functions altogether in the relevant categories from our heuristic analysis and tree test. One question was “Let Mitski know you really like their song” and that was misperceived by almost every user and could’ve been specified towards commenting specifically, rather than relying on a question asking if they ended up commenting a couple questions after. On the other hand, the idea with the “Follow” button feeling non-important was brought up from a question on how users would support an artist, which ended up being incredibly valuable despite not focusing on that specific function. So although specificity is valuable, it can bring up deeper thoughts that participants might have, but I would still spend further time on creating these questions.

Even during the A/B test as results were coming in, I learned a firmer perspective on the types of testing needed for specific questions. During our heuristic analysis, we discovered a violation with the visibility of the play bar. We translated this to ask users to identify and turn down the volume, but we could only see the heatmap of clicks, and most of the data we should’ve gathered would be with eye tracking. Most users were able to click on the right spot, but we miss the necessary data to see where they looked first; their initial expectation, so the success rate is high all around, no matter if it was the base website or our prototype.

After presenting our data to the class 3 seperate times, although we improved each time, we had more data each time as well. I want to learn more about how to share our data in a fast and meaningful way to people who may not want to full story. I’m currently still working through that as I’m relaying these case studies I’ve done, but as time moves along, I’ll learn from my peers and mentors to develop the most efficient way of presenting our results to the world.