Refocusing the Company
When our Chief Product Officer Rob Reynolds joined the company, he outlined a three-front approach to focus the company’s efforts over the next two years, which included creating robust course authoring, rich data analytics, and an enhanced social layer. I was proud to be lead designer on course authoring and lead designer on what would become communities. In the fall of 2015, I was tasked to research, design, and refine a large scale initial approach to communities on our platform.
Summer Social Reboot
- Conversation Not Tethered to Content
- See everything in an environment happening in one place
- Build interest around courses not enrolled in
- Build a sense of community for distance learners
- Bridging the Gap Between Professionals and Students
To design types of communities that could provide immediate value to users, we wanted to reach out and analyze social networks that people were already engaging on. We researched mainstream social networks like Facebook, Twitter, and Google+ and more analogous networks like Quora.
At the start of the project there was a ton of divergent exercises, such as random entry to create new ideas. Towards the end, we needed a way to bring the work together and started meeting at lunch and the end of the day to review each other's ideas. This allowed us to converge on similar interaction patterns and provide valuable input on the challenges we were facing.
Interaction Patterns We Were Interested In
- Manipulation of the Activity Stream
- Information Architecture & Behavior of a Post
- Composing & Sharing Content
- Tagging, Mentioning, Inviting People
- Joining or Subscribing to Content
Being a recent graduate, I have a fresh recollection of how social media has intersected the learning process. Still having access to many of the communities, groups, and hashtags I have been a part of over the years, I surveyed how students and faculty have coopted the functionality of large social networks. I analyzed the context and behavior of members in attempt to recognize patterns. It is particular interesting how user to address their needs and influence their engagement.
Insights from Online Observation and Student Interviews
- When Groups are Larger, Engagement is More Formal and Less Frequent
- Nested Communities are Flattened Online
- Temporal & Long Term Engagements
- Community Graveyard, Students Reluctant to Let Go
- Becomes Primary Form of Communication
- Difference between Student Generated & Faculty Generated
Concept Modeling & Low-Fidelity Wireframing
Onion Model Throughout the process I was fascinated about the idea of what I called an onion model. I had been discussing with students about their experiences using social media as a tool for school, when one of them told me that she never realized that the her school had a Facebook page. She had often wondered how everyone knew about upcoming events and why she always seemed to be the last to know. I asked her if she followed the college page and she said yes. As I was observing online interactions between students and college, I realized the problem. There is an inherent nested hierarchical structure to universities that was being flattened online.
I tied this into my own experience when I was attending university. There were at least 6 entities related to my area of study and each of them had their own disconnected Facebook page: University of Oklahoma (University), Weitzenhoffer Family College of Fine Arts (College), School of Art and Art History (School), Visual Communication Design (Major), Graphic Design Association (Student Organization), VisCom15 (Class Cohort). I also found similar issues with other college social media accounts across the OU campus. I wireframed and did several information flows attempting to realize this model.
“The notion of choose your own adventure, used in the video game industry, is a fresh approach to education and presented many possibilities.”
Dynamic, Analogous Interest Our Chief Product Officer Rob Reynolds often writes about the idea of creating pathways for students to get lost in their education. At Next Thought, we needed to do more to encourage instructional design that gave opportunity for student to have more ownership of their education, because that is when students advance the most. The notion of “choose your own adventure” used in the video game industry is a fresh approach to education and presented many possible manifestations. One manifestation I explored from this mindset was inspired by Pinterest’s approach to interest inception. In Pinterest, the user follows one interest to another to another until the original pin clicked on, is forgotten. This approach used content as a node to dynamically derive topics that are aggregated together in a single feed topped by other “interest tags.”
Community as a Non-Linear Course Another take on the “choose your own adventure” mindset was around the idea of freeing content from a sequenced approach. This model leveraged the modular nature of our platform object: assignments, readings, video sets, discussions, interactive timelines, and external links. The idea was that these objects could not only live in a course, but they could also be shared out as a dynamic objects inserted into a post object and live freely in a feed. This more organic approach to consuming course material was particularly ideal for professional development environments where its was natural to be consume some material and let others stream away. This also has the hook benefit of giving users a fear of missing out that would potentially increase check-ins.
Channeled Conversation When thinking about massive communities, on the scale of a university or college, we thought about how activity streams could be segmented in meaningful ways. As users map out possible channels, streams are entirely dependent on the community’s size, specificity, duration, audience, purpose. That truly is the power of the channel model. Any given community can have its own channels, some that anticipate usage and others that respond to usage.
Making Tough Choices
I sat down with our Head of User Experience Aaron Eskam to discuss the explorations I had done up until this point. We already knew going into the project, that this was going to be a huge undertaking for our development team to produce on a narrow timeline. So, we needed to make hard choices about what could be built in the timeline that would be high-impact and low risk. Activity cards were already in use under the course activity tab. So, if we scaled back our approach to just a manipulation of the card streams, and did it in a flexible and powerful way, we would have a short term win. This reality check gave way to us reaching a consensus that the channel model made the least assumptions about communities and provided a powerful way for them to adapt based on context. It was not about originality; It was about impact.
Click Through Prototyping for Key Stakeholder to Review
Even with scaled back approach, the work was far from over. I needed to work through the user flows and produce a polished click through prototype that exhibited most of the interaction and behavior to show Rob.
In that meeting I could see Rob, who was a full time educator, thinking through all the possibilities and limitation the solution afforded. After a period of pause, he expressed a good deal of excitement at what this would mean for the company and how we would use it going forward.
Everything happened really fast after that. Aaron and Rob went to present it to the Provost and Board of Directors at the University of Oklahoma, which has been a close partner with NextThought since the beginning. I started finalizing the UI and porting the experience over to our mobile web application.
Design for the Worst Case Scenario
The design of platform communities was intended to have user input, meaning that clients would take an active role in guaranteeing its success. I made the assumption that they would be eager to provide community descriptions, cover photos, and have ideas about what channels would be appropriate for their context. Unfortunately, this was the assumption not the rule. In fact some clients did not want their communities activity fractured between channels and requested they all stay in the same tab.
It is Not Over Until Launch
Communities was the first flagship multi-state, multi-platform, interaction dense launch that I was involved in. It would be fair to say that I did not realize how much of a proactive role design must take during the development process to make sure the product stays on course. Looking back, we sprinted for close to a couple months trying to launch communities and I do not think I would have pushed harder than I did.
But, I do have a better appreciation for the transparency and priority that comes with integrating yourself into the quality assurance tool that the development team is actively using. At one point, that communication tool was Trello for our team and it is now Jira. Going where your developers are and inserting unexpected interactions and visual design bug tickets, the same way they document server errors, ensures that design concerns are captured, tracked, and fixed. Even if it gets a medium or low priority, it is in their queue and in my experience since adopting this habit many will not forget about it.
Planning for the Day After Launch.
Communities need constant support and in many contexts, that does not always come in the form of a UX or UI solution. Many times, that requires a human component. Communities need a person to initiate and encourage engagement. Communities cannot thrive solely on consumption; you must have content creators. Without a steady flow of original content, communities will not succeed. Our experiments with communities across multiple clients in both professional development and higher education, reveal that when a strong actor is leading engagement people are willing to participate and invest in the exchange of ideas. However, when there are not strong actors, communities are neglected.