Phase I: Understand
User Research
How Might We enable more natural social interactions in a world shaped by COVID-19?
This broad HMW guided the questions we asked in our survey of incoming freshmen to better understand their pain points so we could tailor our design to their needs.
Research Methods
• Surveyed 180 students
• 8 semi-structured interviews
• 2 focus groups
• 6 competitor analysis
Problem Statement Revisited
How Might We help incoming freshmen acclimate socially to a campus with social distancing enforced?
• HMW ensure a smooth and seamless social orientation for new students?
• HMW help foster meaningful relationships virtually?
• HMW re-capture the chemistry of in-person interactions?
Competitor Analysis
We carried out a competitor analysis by scouring the internet and looking at reviews of various solutions to find user pain points. We downloaded existing apps to put ourselves in the users shoes and identify opportunities.
Phase II: Define
Target Audience
After a range of formal and informal conversations with over 30 freshman about their needs, motivations and frustrations for virtual college, we came up with and refined two types of personas.
1) Derek, would like to a more professional and structured opportunity to connect with other students who study the same subjects as hime. He is looking to network with students and professors and is introverted initially but bonds with others after doing a shared activity (working through a problem set together)
2) Niki, is fun-loving and a people-person and is most excited to meet people from different backgrounds in a more social and casual setting. She’s very concerned about making friends given a virtual college experience.
By defining our users journey we could narrow our solution and eliminate features that would confuse them or complicate our platform.
Phase III: Sketch
Lo-fi wireframes
Though we had planned for an “events” and “communities” page, we went for a much simpler solution and settled on the matching form as our key focus of the platform.
Phase IV: Prototype
User Testing Plan
Before our prototyping our platform we came up with a couple of design alternatives and  planned user tests to determine which platform/existing solution worked best. We came up with a robust user testing plan keeping the number of people, length of interaction and the platform variable and all other factors (hour of day, facilitators, icebreakers) constant. We had 2 focus groups and asked our users to fill out a  later, rating their experience on the platform and their enjoyment of the interaction. Once we had planned these user tests, we were ready for execution.
 People had a really positive experience talking to other freshmen in groups of 4-5 over a non-zoom video conferencing platform for 30-50 minutes with guided questions
• Everyone wanted to connect again
• People enjoyed interacting through icebreakers
•Wanted to meet once a week
• Some wanted random matching and others based on interest
• All students said they were very likely to recommend this toa friend
• Our most popular events were: AMA, game night, chill/chatty sessions
Phase V: Iterate
Styleguide + Branding
We wanted a vibrant and fun theme. Something attention grabbing, yet clean and easy to read. Since our users were Gen Z users, we decided to opt for an MS Paint theme, kind of tongue in cheek. Our graphics contained plenty of pixel art and our copy was always witty and humorous so we could be relatable and make the incoming first years feel comfortable.
Launch
Part I : Pre-Bondings
Students sign up on our website to get matched. Only those with a Columbia UNI can sign up to maintain privacy and exclusivity. Students will be asked to fill out a short questionnaire which will be based on the psychology of intimacy. They can sign up for our newsletter and follow us on the gram to get a sneak peek of our community.
Students are matched in groups of  eight based on sharedinterests/backgrounds. Students will be asked to fill out a when2meet stating their availability for the week. Calendar invites will  be sent later with a link to the virtual room hosted on which allows students to hop in and out of other rooms and see who is in which room. This ensures students don’t have to stress about planning the meeting.
‍
Admins can set limits on the number of people in a room. A Connector will be present to facilitate conversation initially until students get more comfortable and can go into their own room. They are given open ended fun questions based on party games and NYT’s “36 Questions to fall in love". The question will be from a randomized list and will be open-ended enough to allow for broad discussion but limited in scope to draw out clear answers.
These questions will ultimately serve to enhance connections and create lasting friendships. The students will be invited to 3 of these bonding sessions, each with a different theme like "game night”, “talent show” and “present your partner.” Repeated interactions breed closeness and familiarity.
Additionally, there will be “Happy Hours” every Sunday at 12pm on the platform where all are invited to virtual rooms. This will allow students to meet others outside their bonding groups and will be random in that no one knows who will come to which room. (Like Omegle, but for Columbia Students only.)
Part IV : The Wider Community
A newsletter and Instagram with engaging content like polls and “Live Q&A” and “Ask Me Anything” will further create a sense of community.
Reflections + Challenges
Working with a small team of just three other people across various timezones posed its own set of challenges. But more than that, scalability served to be our toughest challenge. We expected around 100 people to sign up when we first launched. But over 400 people signed up. This was more than we could handle and we quickly had to arrange for more Connectors at short notice. We couldn’t do everything - so we had to hire a social media manager and a graphic designer. We had to draw up budgets, have meeting with other start ups interested in collaborating and constantly collect feedback and incorporate it into our design. Timelines changed based on external events (like our school announcing it would be completely virtual rather than the original hybrid plan) This is perhaps why Columbia Connect grew so rapidly and became so popular.
In terms of design challenges specifically, it was tough to come up with a digital solution that wasn’t another Facebook or Instagram. We wanted the focus to be on quality of interactions rather than quantity of users. People had differing opinions on the platform features and it was difficult to separate the effects of the moderator (and other external factors) rather than the platform itself. We had to control our variables and stick to the script during our focus groups. If we were to do this all over again, we would come up with more robust and scalable user testing plans.
Overall, it was so fulfilling to see an idea become a sketch to finally a real life platform which made a real impact in peoples lives. Winning the Social Interaction Design Challenge was a high point, but reading those emails from first years who told us they met their close friends thanks to Columbia Connect? Priceless.
