← All projects

UX Research Proposal

Audio
Streaming
Preferences

Why is Spotify a preferred audio streaming platform, and which of its features are most popular?

Role
UX Researcher
Team
Group of 6
Duration
2 weeks (August 2022)

Collaborators

Caroline Nielsen · Dylan Kapadia · Eric Harrison · Katie Teitelbaum · Marissa Villano · Edward Chase

INFO 690 — Understanding Users: User Experience Research Methods

Goals

  • Learn about users' audio streaming service preferences
  • Learn about users' audio streaming service usage and habits

Background

The Platform

Spotify has changed the way the world interacts with the music industry.

  • Founded by Swedish web designer and entrepreneur Daniel Ek, with Martin Lorentzon
  • Released October 2008, after Ek set out to tackle piracy in the music industry
  • Reached the United States in July 2011
  • Two million paying subscribers by October 2011
  • Today more than 207 million users across 65 countries

The Research Gap

Popular audio streaming companies generally provide their users with similar services.

  • Previous research has determined which platform features users enjoy
  • There is minimal research on why Spotify specifically is deemed a top streaming service
  • Accordingly, we plan to investigate what differentiates Spotify from other platforms

Why It Matters

There are many reasons the findings can benefit Spotify, both the product and the company.

  • To assist Spotify experience designers and application developers in focusing on the appropriate platform features they need to enhance
  • To direct developers to potential new features that will benefit users as well
  • To improve user experience and Spotify's competitiveness

Proposed Methods

Two methods were selected to sufficiently conduct the study and obtain accurate results. The second runs as a follow-up to the first.

  1. Digital Survey
  2. Task Observation + Diary Study

Method 01

Digital Survey

An online survey whose goal is to collect data on participants' sentiments.

Participants

  • Targeted participants:
    • Located in North America
    • Born between 1981 and 1995
    • Have used or currently use Spotify as a streaming service
  • Since Spotify collects birthdates and geographic location upon registration, this backend data can be used to scope in-scope users

Recruitment

  • Current users:
    • A non-intrusive in-app pop-up to learn more about, and register for, the study
  • Previous users:
    • Utilization data and stored contact information identify users inactive for at least 45 days before launch
    • An email campaign acknowledges their usage cessation and politely requests feedback through the survey link
  • Both groups are routed to an overview page outlining the study's purpose, expectations and length, then to registration where they:
    • Are presented with informed consent information on potential risks and benefits
    • Acknowledge the information and verify that they wish to continue
  • Anonymity is made clear to participants up front, to decrease possible response bias
  • Participants then receive one of two surveys, depending on whether they arrived from the pop-up or the email campaign

Sample Sizing

North America holds roughly 87.7 million Spotify monthly active users, 47.9 million of them premium subscribers (Dean, 2021). A sample of 1,000 is large enough to account for the variability and diversity in behavior across a population that size.

Composition of the 1,000-participant sample by segment and recruitment channel
Segment Recruitment channel Share of sample Participants
Current users — non-premium In-app pop-up 25% 250
Current users — premium In-app pop-up 25% 250
Previous users Email campaign 50% 500
Total 100% 1,000

Most people invited will not respond, so recruitment starts far wider than the target. We assume a conservative 10% response rate from the pop-up and 5% from the email campaign, where old addresses and spam folders add engagement barriers, and so plan to approach 5,000 current and 10,000 previous users to land these completes.

Method 02

Task Observation + Diary Study

A combination of task observation and a week-long diary study run as a follow-up to the survey. The goal is to analyze how participants use Spotify, bridging the gap between perception (Digital Survey answers) and reality (Task Observation and Diary Study).

Rationale

Where the survey provides data on participants' sentiments, this method allows us to analyze how participants are actually using the application. It lets us see the path taken to complete tasks, and gather data on how participants interact with Spotify when not being observed — a much more organic picture of how the app is used.

Recruitment

  • Same population as the Digital Survey
  • Post-survey option to opt-in to further studies (current users only)
  • Total sample size of 50 participants, randomly selected

Ethics & Risk

  • Participants receive an informed consent agreement covering purpose, potential risks, benefits and any other areas of concern
  • Upon evaluation, the research team was not able to identify any ethical concerns
  • The main risk is the Hawthorne Effect:
    • An observed participant may fumble tasks that are otherwise done without issue
    • This could skew analysis from a seemingly uncomfortable interaction with the interface
    • Mitigated by pairing observation with the diary, capturing unobserved usage
  • The benefits far outweigh the risks, chiefly an understanding of what makes Spotify preferred over Pandora, Apple Music and Google Music

What Participants Do

Participants first work through five short observation tasks covering their library, liked songs, search, shared playlists and podcast discovery. They then keep a diary for one week, answering the same eight questions each day about how and how long they used Spotify, which features they reached for, and whether they turned to any other platform.

See the full tasks and diary questions →

Comparison of the Methods

Both methods provide strong stand-alone results; using them together provides greater depth and analysis. Each method's weakness is the other's strength.

Strengths and limitations of each proposed method
Method Strengths Limitations
Digital Survey
  • Effective at gathering preliminary data
  • Reaches thousands of people in a fraction of the time the second method requires
  • Volume makes it possible to find raw trends and patterns
  • Large volume of data comes at a potential cost
  • Many biases can come with a survey
  • People may not pay attention to the questions
  • People may think from a different perspective than their actions show
Task Observation + Diary Study
  • Visually presents what participants actually do as they click through tasks
  • An organic method that provides candid data on how people use the app
  • Drives meaning behind the survey answers by answering the "why"
  • Awareness of being observed may alter results and skew data
  • Takes more time and resources than other methods
  • Reaches far fewer respondents

Proposed Timeline

Both methods on one relative schedule. They share an initial research phase and a closing analysis and write-up, and Method 02 adds the observation and diary week.

Month 1 Month 2 Month 3 Month 4 Month 5 Month 6

Method 01 · Digital Survey

Conduct initial research 1 month
Finalize questions 2 weeks
Distribute pilot 1 week
Recruit participants and collect responses 1 month, or until enough have participated
Review and analyze results 2 months
Construct a journal to announce results 1 month

Method 02 · Task Observation + Diary Study

Conduct initial research 1 month
Finalize diary study questions 2 weeks
Distribute pilot 1 week
Recruit from Method 01 participants 1 month, or until enough have participated
Observation and daily diary, then collect results 1 week
Review and analyze results 2 months
Construct a journal to announce results 1 month

Durations are relative, with no fixed start date. Recruitment windows may run longer if participation is slow.

Conclusion

Streaming services have grown in scale and popularity over the last ten years, changing how songs and musicians are marketed. Spotify has contributed to that change, becoming one of the most popular services in the market. This study asks where the value in Spotify lies for its users: what makes it stand out, and which features are part of that enduring popularity.

Both proposed methods are viable. The survey obtains more detailed answers from a larger number of respondents and helps explore the "why." The observation and diary study gathers more concrete data from fewer respondents, with some risk of altered behavior from being monitored.

Recommendation

Conduct both methods. Together they cover each other's weaknesses while keeping their strengths. This costs more and takes longer. It also gives the best result. Spotify could grow as a service, offer a better experience to all users, and push other audio streaming services to improve their own products.

Research Instruments

The full survey, diary and consent materials proposed for the study.

References

  1. Baxter, K., Courage, C., & Caine, K. (2015). Understanding Your Users: A Practical Guide to User Requirements Methods, Tools, and Techniques (2nd ed.). Waltham, MA: Morgan Kaufmann.
  2. Dean, B. (2021). Spotify User Stats (Updated Oct 2021). Backlinko. https://backlinko.com/spotify-users
  3. Lidsky, D. (2018, August 6). The definitive timeline of Spotify's critic-defying journey to rule music. Fast Company. https://www.fastcompany.com/90205527/