Parea App
Mobile app that helps busy adults find and make new friends through hobbies and interests. The goal was to design an app, using the Lean UX framework, taking a concept from an idea to a high-fidelity app design.
Lean UX, academic
8 weeks
Researcher, UX Designer,
UI Design
Challenge
We had trouble testing the pairing concept in an ideal way due to time constraints. Too many designers in the kitchen led to disagreements, overlooking important elements our users need and not putting enough thought into information architecture. All of us were juggling a busy class schedule, life and jobs. Communication was key for us to get through these hurdles. I learned that effective discussion, documentation and assigning evenly split up tasks is ideal to making sure everyone is on the same page.
Recap
The idea, proposed by our team lead, was to help create a platform specifically for finding friends through mutual hobbies. Our target users were busy adults, those new to a city/area and those who have been socially isolated due to the pandemic. We learned that everyone we interviewed and tested our MVP with was excited and extremely enthusiastic about the concept of the app.
Process
Project Kickoff: Because this was a school project, we came up with our own business objectives and assumptions. We defined the business domain for our app idea, what competitors have primarily focused on, what existing products/services fail to address, how we will address those gaps, and what audience segment our initial focus will be. This helped us formulate our business problem statement.
Sprint 1, Week 0: In Week 0, we focused on business outcomes and defining how we'll know that we have solved the business problem and how/what we would measure. Because this is a school project, we couldn’t test or measure for success. Next, we defined what types of users we would focus on first. We did this by creating two proto-personas, which is a persona based on the team’s collective assumptions.
We then wanted to understand what our users would be able to get out of our product and the behavior we could observe that would tell us they achieved their goal(s). After that, we ran a design studio with two rounds of ten minutes each sketching ideas that we thought might solve our business problem and meet the needs of our users at the same time.
Combining all our assumptions into Hypothesis statements, we took those hypotheses and placed them onto a prioritization canvas to understand the risk vs value to the business. This was hard to do as we were still gaining an understanding of our product, but we gave it our best guess based on what we know of current apps in the same or similar domain.
After that, we made a hypothesis table and placed each in order, from riskiest to the least risky. This is what allowed us to identify what we would build, ideate on how we would build an MVP (minimum viable product) of it and test first in the next sprint week, creating a backlog of less-risky hypotheses for the sprint weeks that would follow.
Sprint 1, Week 1: This is when we started to test our MVPs and conducted standup meetings every two to three days. Our two riskiest hypotheses involved invites that match people to activities and events based on their profile (location, availability, hobbies, etc.). The other involved customizable notifications pairing people with each other (like the Slack donut bot). These were the riskiest because we identified these assumptions to be what would cause the entire product idea to fail if they’re wrong.
We brainstormed what the quickest and easiest way would be to test our assumptions. We recruited three participants, and sent them a survey form asking their location, availability and their hobbies. We then followed up by sending them an invite to a fake event, using a Canva event flyer template, and customized it to match their interests.
We interviewed each participant to get their feedback on the invite and dive deeper into what they were feeling, doing and thinking when they received the invite. We also asked them questions to uncover their behaviors and current pain points when it comes to finding and making friends. After all three interviews, we did affinity mapping as a group to find common patterns and insights. We learned that our interviewees liked the event invite and found it relevant to the info they provided about themselves.
For the other assumption, we had a hard time replicating this same experience of pairing people together based on their location, availability, hobbies and interests. This is because we had to recruit and coordinate our schedules with the schedules of our participants, leaving us very little time to manually pair strangers together, wait until they chatted or met virtually in a video call and then gauged their experiences in a follow-up interview.
Simply put, it was too difficult and time consuming. Instead, we sent a fake pairing invite, after each participant filled out the same survey form, that showed a notification and fake information about the fake person they were being paired with. It was a scrappy and fast solution, but it worked because we were able to test the idea with all three participants, all of which said they liked that it was customized to them and that they would engage with the pairing invite by sending a message to the person first.
Sprint 1, Week 2: We discussed what we learned from in the first week which was that our assumptions were correct. So, we continued to build onto our existing MVP’s by polishing the event invites to include more info our participants said was important to them such as the date, time and address of the event. We also improved the pairing notification we sent them by making a second screen that allowed the interviewee to see more information about the fake person they were being paired with, allowing them to see their full profile and see all that they have in common. This proved successful as we received positive feedback on both MVP’s. We continued to ask questions in our interviews that gave us a deep understanding of what qualities people look for when making friends, what are the biggest hurdles for them to find and make new friends, what their schedule and work/life balance is like, behaviors and attitudes around small group events vs large group events, talking with new people, how they prefer to communicate, etc.
Sprint 2: We revisited our assumptions, personas, and hypotheses.
Sprint 2, Week 0: Using what we learned from the first sprint, we revisited our assumptions, personas and Lean UX canvas to evaluate if we needed to make any adjustments and revalidated our product backlog. The changes we made to our personas included details about communication preferences, safety preferences, and behaviors towards finding and making friends. Group preferences, small or large, were one of those behaviors. Some people prefer a larger group setting, depending on the event type, and others prefer smaller groups regardless of event type due to their personality (introverted).
Sprint 2, Week 1: We focused on testing a scenario walkthrough of our UI with interview participants. We continued to start each interview off with questions about them, their thoughts/feelings around finding and making new friends, and other important questions. We would then have them test a happy path flow of the two core features of our app, the pairing invite and event invite, by giving them tasks to complete. We had already built an MVP of this early on so in this sprint, we refined those screens and added more such as the log in screen, onboarding screens, a profile page and a notifications screen to cover our happy path flow. This helped us further test our core features with new participants, gauge feedback and fix any pain points/areas of confusion.
Sprint 2, Week 2: In this final sprint week, we continued testing our core features and all other screens we added with three final participants. With feedback, we polished all screens and added anymore that we needed to complete the app. Decisions about the color palette, imagery, icons and the copy also took place. We also had to figure out the information architecture, deciding where content should be placed within the app. We used user feedback and team discussions to do make these decisions.
View Prototype
Conclusion
We were unable to measure success using KPI's and business objectives, however, we did follow the entire Lean UX process from project kickoff to prototype delivery. I learned that Lean UX is a powerful design method that can be used for balanced teams. I really like how the process encourages testing assumptions early and often, using MVP's that take the least amount of time and effort to build. It also emphasizes the importance of UX throughout by keeping user needs and goals at the forefront for the entire team.
We faced challenges with being able to build an MVP for our second riskiest assumption, the pairing invite, but were able to come up with a second-best MVP to replicate this experience and test with people to validate. We also faced challenges in making design decisions. This was a result of too many chefs in the kitchen but we were able to get through disagreements by communication and explaining the reasoning behind our rationale.
If given more time, I would have: come up with a better branding and design system, to test one of our other assumptions, test the final UI with more people to get more feedback and improve the UI (different color palette and different icons).