SoundHound Personas
September 2019
SoundHound is a song identification app that listens to a song and quickly tells you what it is; it can even tell you what song you’re singing or humming. Because the app does not collect any identifying data, it was unclear who the app’s users really were. Tasked with creating personas, the 2 person UX Research team surveyed and interviewed users to create personas to aid design and marketing. Stemming from the personas, further research was deemed necessary and occurred after my internship ended.
Background
SoundHound is a very successful app with over a million users, but we had little information on exactly who those users were. The app doesn’t require you to create an account to use it, and even then, the only information it collects through the account is your email and name. In order to learn who used the app, we decided to create user personas to further inform design and product decisions. I interned at SoundHound, Inc. during the summer of 2019, and this project spanned that entire time.
Process
As I had just begun my internship after the kickoff of this project, I worked under the direction of the Senior UX Researcher for some of it, and later I took full ownership of some steps which I will point out when relevant. Being that this research would lay the basis for future design and product decisions, we allotted a lot of time to properly design the study and analyze the data. We conducted all the work over a two-month period.
In creating the personas, we had three main research questions:
Who are SoundHound’s users?
What motivates and challenges them?
What are their technology and music habits?
As creating personas was new to both the Senior UX Researcher and I, we did some background research to understand similar companies’ processes. As Spotify is also a music app with a detailed & published record of their persona process, we mainly looked to them to get an inside peak at where we should begin. Equipped with that research, we created our test plan.
First, we implemented an in-app survey using GetFeedback.com to gather user demographics and habits related to our app and to music listening in general. I analyzed that data with SPSS, and we decided to narrow down based on user age. Determining which users to contact was difficult since we couldn’t really rely on usage data to determine which types of users to reach out to, but we determined that relying on self-report to identify power users was our best option. Using the survey as a sampling method, I contacted interested participants for an in-depth interview via video chat. After many rounds of recruitment and emailing back and forth with participants, we regrettably had a somewhat small sample for interviews, but due to budget and time constraints we had to make-do.
After consulting with the other researcher, I developed a set of questions to ask during our hour-long interview sessions. We ran a couple pilot interviews to see how it went, and from that, we reassessed our questions to better suit what we were looking to learn. After observing the first few rounds of interviews, I began to conduct the interviews myself and truly enjoyed the opportunity to connect and learn about these people.
With all these steps behind us, the Senior UX Researcher used the prepared data to construct the final personas. She then passed off the information to me, and I designed the visual persona cards that were presented to stakeholders from design, product, and marketing.
Results & Impacts
Finally, we had 3 personas that we determined were our power users. The research highlighted users’ tech habits, their preferences within SoundHound, and details of their goals and challenges in life. After presenting, the personas sparked an important conversation of how to connect designers directly with data and facilitate their own successful use of that information. Interestingly, the personas also highlighted a need for deeper communication at the beginning of a project to determine what deliverables designers expected. Marketing used the personas to target their messaging and to discover potential untapped user bases. Whereas, the designers felt the information that would be useful to them was not reflected in a persona. To them, it was unclear how they could use this information to affect their design decisions. After conversation, it was determined that the next step would be to create user journeys for each persona. Journeys would clearly lay out how the app’s features could best be designed for a better experience.