Roommates began as a mere idea between two roommates who worked and lived well together. Though our own living situation was unproblematic, we recognized through conversations with our peers a problem space that was unaccounted for: how to organize and communicate with your roommates. We wanted to design a product that would improve the shared housing experience for college students by addressing problems related to move-in, chores, expenses, and house rules. However, before we could design an application that people would actually use, we first had to narrow down our problem space and better understand the users.
01 — Existing Platforms
Our user-research began with researching existing platforms that were similar to what we wanted to achieve. The current market has many applications that tackle a single aspect of living with others; for example, only chores, or only splitting expenses. We discovered one app, RoomMate, that all-encompassed the features we had in mind, excluding how to track move-in information. As part of our research, we downloaded this app and noted some key findings:
- The target audience is primarily European users.
- The app has poor UI/UX (the interface is extremely cluttered, the user flow is not intuitive, the design is not appealing)
- We stopped using it after one day.
Why did we end up not using an app that was so similar to what we wanted to accomplish? Although the poor UX contributed to its failure, it became evident that the living problems we personally encountered were not painful enough to change the system we already had in place. After speaking with our mentors, we realized that in order for this product to be successful, it would have to be 10x better than anything that currently exists.
02 — Survey
The next part of our user-research was an attempt to identify the biggest pain points users experienced when living with someone else. Our intuition was that if we can find the largest problems and create a product that targets them, people would be inclined to use the product. We began by releasing a survey that took an initial read of the challenges users faced with shared housing. We asked users to indicate on a linear scale (1 being not stressful, 5 being very stressful) their experience with the following tasks:
- Choosing how to prorate rooms by size/view/etc. with housemates.
- Creating and/or sticking to own house rules (e.g. guests, chores, quiet hours, etc.).
- Tracking which items your house needs (e.g. pots/pans, toilet paper, other items you can share) when you first move in.
- Splitting expenses for household items (e.g. WiFi, utilities, prorating, groceries, etc.)
- Splitting and scheduling chores.
In addition, we asked more open-ended questions regarding how they communicate, what other challenges they might face, what existing methods they use to navigate those challenges, and what solutions would be helpful to them. As a final question, we inquired about whether they would be willing to discuss their answers further in a brief interview.
In total, we garnered about 35 survey responses. We knew to take these responses with a grain of salt, especially since we realized we had asked a leading question (i.e. “What solutions or platforms would be helpful to navigate [your challenges] further?”). However, the primary intention of this survey was to gauge what points were more painful than others — not quantify how painful they were. By looking at the relative ratings of each task, we learned that “splitting and scheduling chores” was the most painful, followed by “creating and sticking to your own house rules.”
A preliminary analysis of our survey results made us realize that our current target audience of Cornell students was too broad. We decided to narrow the audience to Cornell students living off-campus. This audience, however, was still too large, as it included both off-campus apartments/houses as well as cooperative housing and Greek life housing. After weighing the pros and cons of the business-to-business (co-op/frats/srats) and business-to-client (apartments/small houses) marketing strategies, we decided to limit the MVP to only off-campus housings of 2–7 people. We also decided to change the course of the project; rather than rushing the research process and creating a mediocre MVP, we instead are focusing on the user-research and creating a design we are confident users, including ourselves, would use. The most important result of these surveys, however, was identifying users who were willing to interview with us and further discuss their responses.
03 — Interviews
Out of the people who indicated interest in interviewing, we contacted those who both fit our new audience and had more extreme survey results. Prior to the interviews, we wrote some guiding questions based on their survey responses, such as asking them to expand upon their reasoning for rating a task as stressful or not. However, we tried our best to maintain a casual, confidential conversation and let the interviewee guide us through their pain points.
We conducted 9 interviews with users from our target audience. Although we were initially intending to do more (i.e. 12–15), we saw that the main themes of the interviews were converging. Because we started to see repetitive information, we decided to not conduct more interviews and instead focus on extracting insights from the existing ones.
04 — Insights
For each interviewee, we extracted 9 key insights. Using the digital platform Miro, we categorized these sets of insights by their common themes. We determined there to be four overarching themes, each with their own subcategories:
- Communication
- Group chat
- Good communication vs. bad communication
- Repetition/little things
- Confrontation
- Reminders
2. Personal
- Different standards/needs
- Shared/personal belongings
- Initial systems
3. Costs
- Splitting costs
- Prorating rooms
4. Chores
- Chores systems
- Chores tensions
Then, we summarized each subcategory into their most significant points and extracted any suggestions or wants the interviewees had indicated. These summaries allowed us to extract the four biggest insights that comprise our problem space:
- College housemates/roommates struggle to manage and organize shared spaces, primarily due to different cleanliness standards/lifestyles and a lack of communication, established systems, and accountability.
- Creating a system that everyone agrees on can be difficult (chores, COVID, prorating).
- When housemates do set initial agreements (chores, COVID, etc.), the system often fizzles out over time or only some stick to them.
- Current solutions are not centralized (venmo, splitwise, spreadsheets, text, etc).
05 — Takeaways
Ultimately, our user research enlightened us to whom our product should target, as well as what problems we should prioritize. We narrowed the target audience to Cornell students living off-campus with 1–6 other roommates or housemates. This is because in our interviews, we realized that larger houses (including cooperative housing and Greek life housing) operate much differently than smaller houses. Rather than creating a product that catered to both, we decided to limit it to smaller homes in order to focus on the supporting features.
Our research also indicated that the product should be a centralized solution to our problem space. The application should be all-encompassing so that users do not have to switch between platforms, which makes it easier for users to commit and be held accountable. Thus, we intend to design features for:
- Individual & group onboarding (identifying communication methods, house rules, prorating system)
- Chores system
- Collaborative shopping lists
- Expenses
In order to improve the user experience, these features will include automated suggestions or templates (e.g. an automatic prorating calculator, sample chores and house rules) in addition to manual entries. Most importantly, our product will implement a notifications/reminders system for chores and expenses. Because many users eschew confrontation, we intend to create a pinging system, where users can anonymously send reminders to everyone in the house when a chore is overdue.
Another takeaway from our user research is the importance of balancing accountability and negativity. Although the pinging system is anonymous, we know we must ensure that each house and user is given flexibility and the benefit of the doubt. Users will be able to indicate if they are busy and cannot complete a task or chore on time. Additionally, users cannot ping repeatedly nor on the day of when that task was due. Our product will focus on creating positive user interactions through encouraging language in all the notifications and a positive reinforcement system where users can applaud the good work of their housemates.
This user research sprint was essential to the development of Roommates. We began with an extremely broad audience and problem space, which would have been more difficult to develop a product that users would willingly adapt to. However, the insights gained from our user research give us confidence that our product will be a more than adequate solution to the problems Cornell students encounter while living off-campus in a small home or apartment.
