Before starting a new role I worked on this design challenge focused around the Bumble BFF app. I took this design challenge as not only an opportunity to show my skills and abilities, but also to challenge myself.
I was fascinated by the process we took while doing design sprints during my internship with eCompliance, so I used a similar process taking different tools and methods to approach completing this design challenge in just one day.
The following project was NOT done in association with Bumble. This case study was done solely as an educational exercise and should not be associated with the company in any way.
On Bumble, there are three different user types: date, bff, and bizz. In all three of these user types, the profiles stay relatively stagnant. Most of the information collected in Bumble BFF profiles are not relevant to determining if someone would be a good friend or not, and on top of that, most individuals on Bumble BFF don't complete their profile.
At first I wasn’t sure how I wanted to approach this challenge. In reality I'd spend a week or more on a similar problem but I really wanted to challenge myself to tackle this challenge in a day. I came up with a plan and included methods from the Design Sprint process to come up with a few possible solutions for the problem.
I skipped the “Decide” stage as they mostly involved team exercises, and I used the “Validate” stage to outline how I’d go about testing if this were a real problem. Below is a high level overview of the stages and methods I used from Google’s Design Sprints Methodology website.
I made myself a little area with all the necessary tools needed to focus and complete the design challenge. I find that I think better on paper, and like being able to take a hands on approach to solving problems in the early stages of discovering, researching, and ideating.
Our Bumble BFF users are not completing their profiles because they find the information collected on profiles isn't relevant to determining if someone would be a good friend or not.
If we can solve this problem, it would impact our Bumble BFF users positively because they would be able to determine whether someone is a good friend based on relevant information. It would also benefit our business because these users would complete their profiles more often, increasing engagement with the feature/app.
Design a creative solution that focuses on the right information and promotes profile completion for Bumble BFF.
Before starting this project, it was important for me to get familiar with the Bumble app. While going through the onboarding and sign up flows of the app, I made a note of things that stood out to me when completing my profile.
The existing app encouraged users to complete their profile in the onboarding process, but also gave the option to skip and complete aspects of their profile later. Through prompts, users are able to put together a profile that best represents who they are and what they're looking for in a friend.
With this being a design challenge that I aimed to complete in one day, it was inevitable that I'd be making a lot of assumptions. To avoid any personal bias, I wanted to list these assumptions.
With this list I had a clear understanding of what I didn't know, and what I needed to learn through user research.
Users aren’t completing their profiles because they aren’t being asked the right questions to reveal relevant information
Users are technologically savvy and familiar with interaction patterns around popular dating or networking apps (ex: swiping yes or no)
Users choose to complete their profile afterwards but never get around to doing so
The profile creation process is too long and users don’t want to go through the hassle of completing it
Users want to see more information on other people’s profiles
Profile information is a starting point for conversations on Bumble BFF
The "Understand" stage of a design sprint to me is the most important part of any design related project. Without a proper understanding of who I'd be designing for, it would be nearly impossible for me to deliver a solution without any bias. So, how exactly do you go about collecting research and putting yourself in someone else's shoes, in just one day?
The following stage goes through the lightweight process I took to collect research and create some UX deliverables. This process was completed with a hands on approach using whatever supplies I could find around my house, then later digitized.
What better way to collect research from Bumble users is there than messaging them? I swiped through a number of different profiles and kindly got around 4 people to fill out the following survey.
I also reached out to others who weren't on Bumble to get a better idea of how people go about forming friendships, and what they look for in a potential BFF.
The survey brought back a great deal of data, but lacked in detailed responses. With no budget, I looked for blogs or articles by people who have used Bumble BFF in the past.
Through these blogs, I found more detailed writing around how users go about making their profile, and their overall experience with the app.
I bundled the responses into an affinity diagram to make sense of all the data collected. With this exercise, I was able to develop deeper insights around the meaning of friendship and how that translates into an app like Bumble BFF.
A best friend is made over a long period of time, with themes of effort, comfortability, similar interests/experiences, and more playing a role
Users have a hard time expressing themselves accurately through the app whether it be through their photo, bio, or other form of information
There is a feeling of having to “sell yourself” that people find difficult when it comes to making friends at an older age
Moving forward, I wanted to use the findings and research to visually tell a story through the use of personas. In a more common scenario I'd go about making multiple based on varying data.
Since the research I was able to collect mostly consisted of 18-24 year old women, I chose to focus the story in this perspective. With more time and resources, I would have diversified the research process to collect data that better represents more Bumble BFF users.
I used job stories to focus more on the job to be done as opposed to just one user. Although I focused the story on one persona, I wanted to ensure the solutions could still apply to multiple users.
When I am looking for a new friend I want to know if we have similar interests/experiences so I can be sure that we can relate on a number of different topics
When I am looking through people's profiles I want to find relevant information quickly so I can avoid wasting my time on people who I wouldn't mesh well with
When I am creating my own profile I want to make sure it portrays who I am accurately so I can get more matches and attract people who are similar to me
Using a user journey map was the best way to put all this work together to clearly outline any opportunities. Skipping beyond the point of onboarding, I started the journey at the point where one would first get into the app and start looking at profiles.
From there, I created how might we statements that informed the opportunities and ideas I'd later explore in the next stage.
Create a Bumble BFF account to find new friends in her city
How might we allow users to quickly add relevant information to their profile
How might we encourage users to fill out all parts of their profile
How might we provide the most relevant parts of one’s profile to a user
How might we minimize the time it takes to read/look through a profile
The process up to this point had given me enough information to consciously make design decisions that can effectively accomplish both user and business goals. In the following stage, I took the time to more clearly define specific context and desired outcomes of potential solutions. This would allow me to focus my attention more clearly on solving the problem at hand.
For uses to engage with their profiles
Number of times a user visits their profile page
Number of times a user completes a part of their profile
Number of times a user edits a part of their profile
Average % of profile information being completed
Number of updates being made to profile per day/week
Number of visits to profile per day/week
Number of conversions from blank profiles to filled out profiles
Average duration on profile page
It's easy to want to quickly jump into a design tool and put a solution together, but often times I find this approach doesn't allow me to think freely. Trying to rid myself of those constraints, I took a pencil and some cards to sketch any ideas that came to mind. These sketches would later inform the designs I put together in following stage.
I used the Crazy 8's method to time myself and quickly sketch together eight different solutions in 8 minutes. From the eight, I focused on the five that I believed would be the most effective and designed those in a higher fidelity.
From all this work came the following four solutions, which all uniquely aimed to increase profile engagement in different ways. At different touchpoints in the app, the following solutions would deliver a new experience for Bumble BFF users to find friends with ease.
This feature focuses on connecting people based on compatibility. Using various indicators from your profile, we'd connect you with others who would seem like a good match.
To get more accurate matches, one would have to complete their profile entirely. This feature would serve as a motivating factor to encourage users to complete their profile.
Including interests in your Bumble profile allows you the ability to express yourself in new and interesting ways.
While browsing other profiles, if they have interests that you also like, you can simply add those to your profile with just one tap.
Find new interests? No worries, your matches will adjust with you! As you change your profile, your match score with existing matches adjust accordingly.
This plays into the same motivation that comes from the first solution, at a different touchpoint.
The existing Bumble BFF app allowed users to answer prompts that were pre-determined through the onboarding process. In context, these questions feel uninspired and unrelated.
With this feature, users can have conversations with matches and add the responses they send to their questions to their profile. In a way, it is an opportunity to answer frequently asked questions and add something unique to their profile!
With a thoughtful consideration for responsive design, I made sure the solutions I created would work across various devices. Bumble has both a web and mobile app to consider, and so I took some time to visualize these solutions on the web.
Through working on this project I was able to gain a higher level understanding of what my design process looks like. Obviously in a real world setting there are a lot more considerations, constraints, and people to consider when solving a problem like this one. Regardless, I think my approach to this problem was executed well as I fortunately received a job offer after presenting it to the team!
If there's anything I've really learned about myself here, it's that I work well under constraints. I've learned that my ability to adapt to any situation and deliver high quality work regardless of time or resources is something that translates well when working as a product designer.
Often times it can be difficult to get the appropriate budget or resources needed to conduct user research effectively. Under the time constraint that I gave myself, I effectively found interesting ways to collect research to avoid personal bias.
When it comes to planning and executing design sprints, there's a lot of factors that go into determining what activities are the most effective to get to a solution. Giving myself only a day, I had to carefully plan my process to ensure the final result would be satisfactory.
Within approximately 10 hours I went from no knowledge of Bumble to having multiple solutions that would increase profile engagement. While these solutions would need to be tested to determine which is best, I'm proud of my design execution.
This challenge was not based on a real problem, but if it was, I would've approached it much differently. I would've included my team along the entire process, conducted usability tests with users, and iterated on these designs before launching any of these solutions.
To stay within my given timeframe, I only sketched out a couple of ideas and went forward with those that I deemed most likely to succeed. In reality, I would've like to explore more concepts, outline the benefits of each, and work with a team to decide on one or two to take into testing.