Boris is a product designer and researcher working the San Francisco Bay Area

I was the team lead of an 8 person group redesigning Facebook Groups  It was an 11 week long project to create a cohesive, consolidated, and streamlined group experience.  We conducted at least 31 user interviews, drafted 6 versions of personas, held multiple interpretation and visioning sessions, crafted an expansive affinity diagram, created and tested multiple paper prototypes, created 2 rounds of high fidelity prototypes in Axure, and presented our work multiple times to a happy UX minded crowd. All of the team contributed to research, ideation, and prototyping.  I personally had a lot of input in the visioning and ideation stages, put a lot of work into the interviews, personas, and affinity diagram, and created all of the hi-fi Axure prototypes, as well as some of the user flows and info architecture of the low-fis.

The Process

Step 1: Identifying the Problem

This is the board upon which we  started to crystallize the problem which we felt was substantial and challenging enough to tackle. 

Our task was to design a solution to a practical and meaningful pain point, or problem, in a real world process.  Even before we could get together, we had to schedule meeting times.  We opted to do this online.  From there, we thought it would be useful to create a Google Doc as a virtual whiteboard of our ideas.  As our ideas grew, we had to use Trello to keep track of all the information and tasks that were going around.  Halfway through this painful team organization process I realized that getting teams on the same page was a real pain in the neck. None of these services provided the most vital aspect of group work, a platform for communication.  They did not support collaboration, and thus, they did not support teamwork.  I suggested we create a Facebook group for our team and attempt to create a designed solution for the small scale team collaboration process.  The rest of the team agreed that this was a challenging and interesting problem and there we started our project.

Step 2: Scope & Ideation

Initially, we hoped to create a unified platform where groups of users could complete their work in a single shared space. This centralized hub would have included access to common collaboration services such as file sharing, screen sharing, communication via audio/video/chat, scheduling, collaborative document creation and editing, and task management. However, we had not yet been exposed to enough user data when idealizing the scope of our project, and we quickly realized our initial ideas were quite ambitious for an 11 week span.  As we were concluding our first round of user interviews we quickly realized that

1) we wouldn’t be able to create such a large system in such a short time and,

2) that perhaps we were trying to support too many actions in one system.  I quickly realized that we would be better served by finding key areas and natural breakpoints within the process and focus on designing one such natural section of it and how it integrates into the others. We started at perhaps the most important and most apparent section of the process, communication; and vying for low hanging fruit, we chose to expand from and redesign facebook groups, which was according to our usability study and the inherent proof of the users, already a wonderful, widely used platform that supported communication.  We then decided as a team to limit our audience to students, and other informal, small team projects that were using Facebook Groups as a communication tool and hub for group work.

Step 3: Interview, Persona, Affinity

The team, gathered round the affinity diagram, mining for insights and looking goofy. 

Throughout the ideation process and some-time after it the whole teams efforts were concentrated on getting lots and lots of user data and building a crystal clear and actionable understanding of our users and their processes (including tasks and needs) for group work and collaboration. We gathered data in a variety of ways, but mostly through interviews and contextual field studies. In both cases, I elected to have 2-3 team members assigned to an interview or study. This was so that there could be a person that was always focused on asking questions and understanding and empathizing with the user from a first-person perspective and so there could also be at least one person that would observe (and record audio/video) the natural interaction and workflow of the user from the 3rd person perspective.

This approach, which I believe coupled a psychoanalysis with an ethnographic view of the user, proved to be useful in helping us isolate key aspects of communication such as rate of speech in natural conversation, and where it fit into the collaboration process, such as in instances of task negotiation, feedback and review of work done, etc.  Using this approach we interviewed/observed (multiple times) another design team working on a mobile app design and a research team that was a subset of a larger lab that was gathering and consolidating data. We also did long term systems studies of an informal time swap group for a café and the online backbone of a club, ‘habitat for humanity’.   Based on those studies we iterated through a series of personas and finally took about 3 days to chart out our understanding of the users, their process, and their interactions, from beginning to end, on an affinity diagram (which can be seen in the photos above).

Step 4: Prototype, test, prototype, test, prototype…

Armed with our personas and the affinity that was borne from them we quickly set out to make some basic paper prototypes. We did that in about a day and then we proceeded to test them the rest of the week. While testing the prototypes, using the newly generated feedback I set out to build out a higher fidelity version of the prototype for a more accurate and natural testing tool. We kept working on the low fidelity prototypes in parallel to still gives us freedom and rapidity when needed. We iterated through these steps about 3 times and by the end I had built many different prototypes at all levels, some of which were fully interactive.

Step 5: Present findings and redesign

Over the course of 11 weeks I and two other team members presented our project 3 times, with our final presentation showing of our complete findings and our designed solution to date. We presented to our studio human centered design class, which was taught by Jim Hollan, and were  met with very constructive critique and feedback alongside modest praise and mandatory applause. It was a wonderful experience all in all and the hunt for a perfect (or closer to) group collaboration system is still on! Maybe one day I’ll return to the project myself.