In 2008, my team of 5 worked with Linden Lab to improve user retention rate through social networking.For 8 months, we conducted user studies, created designs, and developed a software prototype for a recommendation system powered by current Second Life residents. I was the co-technical lead and was heavily involved in all the phases of the project.

What I did

My Master’s capstone project group worked with clients at Linden Lab to increase user retention rate in Second Life. As the technical lead, I worked with the design lead to define the proposed system’s functionality, and I coordinated with the project manager and user studies lead to ensure timely delivery of each prototype. I was also the webmaster, responsible for creating a website to document our process and results. All five of the team members worked together to conduct user research and create design concepts throughout the project.

To ensure that we devise the most effective solution, we used a comprehensive suite of HCI methods to study existing services and users, experiment with new ideas, and test our solutions.

Learning about the users

Our clients and learned that they want to incorporate some social networking elements into Second Life. We conducted competitive analysis with existing virtual worlds and social networks, then performed extensive user research with existing users of these services, including Second Life.

Through contextual inquiries, we learned that many experienced Second Life users maintain contact with other Second Life residents online and in real life. Meanwhile, our survey results showed that new users have a harder time finding groups and trusting other users.

“How comfortable were you starting the conversation with a stranger?”

Despite this, new users still see social connections as an important aspect of their experience. A novice user participating in our contextual inquiries said:

I wouldn’t be on Second Life if I didn’t find a connection with someone.

Another roadblock for new users, that we found through interviews and think-aloud studies, is the steep learning curve involved in using the UI, especially the search feature, the only feature that aids discovery.

The problem with the new user experience.

Armed with this information, we set out to create a system that:

  • Provides new users with the social resource of experienced users
  • Establishes new users’ trust towards existing users
  • Provides a fast and easy way to discover new things in Second Life

Building a Solution

We brainstormed and tested many ideas using personas, concept scenarios, storyboards, and validation scenarios, and created the idea of a user-rated listing of events, groups, and locations in Second Life.

We initially envisioned an immersive experience  with the recommendation system, but quickly discovered through peer feedback and user testing that users prefer a more lightweight and integrated UI.

We iterated through 4 sets of wireframes and low-fidelity testing.

Our final design consistent of a short toolbar on the upper-right corner of the screen. From this bar, users can see a list of recommendations tailored to their interests.

Comments, ratings from friends, and ratings from people with similar interests provide the user a trustworthy overview of the recommendation.

The user can also rate the current location or event they are attending, and click the “wander” button to be taken somewhere new and matches well to their interests.

We refined the design through paper prototype testing, and built and tested a software prototype with real world users over 3 months. At the conclusion of the project, we provided Linden Lab with the prototype, video sketches of usage scenarios, and complete documentation of our findings and ideas. All deliverables can be download from the project website.