MongoDB

This client project focused on supporting the UX Research team at MongoDB in identifying ways to keep users engaged on their virtual learning platform, MongoDB University. Our findings influenced the product direction, leading to the recently added “Learner Journey” feature.

Role: UX Researcher

Duration: March - May 2023

Team: Two fellow UC Berkeley Graduate Students in the Master of Design program

The Challenge

MongoDB University, a virtual learning platform (VLP) offering free, asynchronous content and certifications for MongoDB software, was facing a challenge. Users were signing up, taking courses, but then failing to return and continue their learning journeys. We were brought in to support the UX Research team at MongoDB and explore this phenomenon in depth.

The Research Question

How might we motivate MongoDB University users to return to the platform and continue their learning?

Goals:

  1. Identify user types and their experiences with MongoDB University and VLPs in general.

  2. Understand user motivations and pain points related to VLPs.

  3. Explore opportunities for incorporating gamification elements to enhance user engagement.

Methodology

The research employed a mixed-methods approach, combining both quantitative and qualitative data collection methods - survey, then interviewing, and last a competitive analysis.

Understanding the Landscape with Survey

We knew we needed to speak to the users themselves. We started with a survey, reaching out to current MongoDB University users and those with experience with VLPs on technical subject matters in general (due to time constraints). While the ideal scenario was to have a perfectly representative sample, we acknowledged the limitations of convenience sampling, utilizing internal university boards, the MongoDB University forum thread, and social media to gather a diverse group of 70+ respondents.

The survey revealed some interesting statistics:

  • 80% of participants valued the flexibility of VLPs, allowing them to learn on their own schedule.

  • However, 75% also reported feeling a lack of human interaction and support on these platforms.

Going Deeper with Interviews

The survey provided a valuable starting point, but we wanted to dig deeper. We selected 10 survey participants for semi-structured interviews, following a pre-defined protocol that focused on user experiences, motivations, and pain points with VLPs.

Note:

We know that proper protocol would require us to create a screener before conducting the survey. Because we were limited on time, the survey data analysis focused primarily on processing data for users that best fit the ideal user profile.

Through these conversations, a clearer picture emerged:

  • Many users, particularly those new to the field, craved the opportunity to ask questions and connect with others on their learning journeys. Asynchronous content and teaching methods made it difficult to build a sense of community.

  • Accountability was another key theme. While some users were intrinsically motivated, others found the self-paced nature of VLPs challenging to maintain.

  • Interestingly, the researchers discovered that more experienced users interacted with VLPs differently. They often used the platform for specific skill development and valued a more streamlined learning experience.

Assessing the Competitive Landscape

To determine how the competitors approach the previously identified user needs, we conducted a competitive analysis. We looked at the six most popular VLPs used by our survey participants:

  1. Duolingo

  2. Khan Academy

  3. Udemy

  4. Coursera

  5. LinkedIn Learning

  6. Unity Learn

We analyzed their approach to motivating users by using the Yu-Kai Chou's Octalysis Framework. This framework focuses on gamification strategies that are scientifically proven to support motivation. We focused specifically on the five elements in the octalysis that rely on positive reinforcement (accomplishment, epic meaning, ownership, social influence, and empowerment) , excluding the three elements that rely on negative reinforcement (scarcity, unpredictability, and avoidance).

Our primary recommendation was to boost the community and personalization scores for MongoDB University to enter a comparable range with their competitors.

We also noted in our findings that the differences in business models may impact these findings but the versions proposed best suit MongoDB University’s model.

Research Findings

Insights:

  • Users value the self-paced nature of VLPs but lack sufficient human interaction and support.

  • Integrating collaborative components could address this need while maintaining flexibility.

  • Accountability measures could supplement intrinsic motivation for some users.

  • More experienced users may interact differently with VLPs based on their goals.

  • Asynchronous content and teaching methods impact users' ability to connect with a learning community.

Recommendations:

  • Level Selection: Personalized learning experiences catering to different expertise levels.

  • Group Quests: Collaborative learning activities fostering a sense of community and accountability.

  • Goal Setting: Features to personalize learning goals and track progress, potentially with gamified elements.

Outcomes

MongoDB University released several new features that incorporated the recommendations from our research. Two specific features are most strongly linked to our research.

The first feature is the Interactive Learner Journey. This feature focuses on the key user insight regarding level selection and goal setting. User are able to jump into a learning path in a way that matches the users individual expertise, instead of starting all learning paths from the beginning.

The second newly-added feature that is strongly tied to our research is the Instructor-Led Training. This feature focuses on the key user insight regarding group quests. Offered specifically to companies, this feature allows companies to facilitate the learning of their employees through customized instructor-led courses.

My Learnings

Our primary challenges faced during this process were:

  • Limited access to internal MongoDB data: We had some difficulty tailoring our research to specific platform usage data as we did not sign an NDA.

  • Scheduling constraints: Balancing team schedules and interview scheduling logistics was difficult.

  • Survey question development: It was helpful to work with the MongoDB UXR team to avoid any leading questions and ensure valid data collection.

However, I gained valuable experience in proper user research methodologies, as well as the importance of building rapport and transparent communication with stakeholders and participants.