Bumble BFF

As a personal project (I have no affiliation with Bumble), I scoped research questions, designed and ran 10 user interviews, and analyzed the data in order to understand how women in their 20s make friends 1:1 and in group settings and a explore a hypothetical group activity feature.

Role

UX Researcher

Company

Personal project

Timeline

July - Sept 2021

Teammates

Research mentor

Research questions

As an avid Bumble BFF user myself (I met most of my Seattle friends on Bumble BFF!!), I have gotten frustrated before with the mismatch between someone's profile and what they're like in real life. I've found myself wishing there was a way to meet more than one person at once, in small groups of 3-7 people, because it seems to be there will be better odds that I could hit it off with someone - and that other people will hit it off with other people!!

To me, small group settings just feel like a more natural way to get to know people than 1:1 settings, which feel reminiscent of a date. But I wasn't sure how other people make friends, and I didn't know what was working and wasn't working for them. Wondering whether other people would be interested in small group meetups, and how that feature might work, I decided to do a personal research project!

  1. How do people make friends on and off Bumble?
  2. What does/does not work well with Bumble and other ways people make friends?
  3. How would group meetups be different from 1:1 hangouts?

Study methodology
Participants

I recruited four participants who were active Bumble users, three participants who were inactive users, and three who had never used Bumble.

Interviews

User interviews are a fast way to learn about users and what they think, feel, and need. I wrote a moderator's guide and conducted 10 in-depth 1:1 structured user interviews.

Key findings
People make friends through many methods

People make friends through a variety of methods - very intentionally looking to meet people interested in new friendships, being befriended by someone else who seemed very intentional, or "just becoming" friends without really knowing how it happened. Most users talked about more than one method, but a few said most of their friendships were from someone else befriending them. Some participants talked about making most of their friends through church, school, work, or sports. Most had "friend groups" from several different activities, and a few participants talked about primarily having "one-off" friends who didn't know each other. Although six of the participants had used Bumble BFF, only three said they'd made close friends on the app.

When no one reaches out first, everyone assume the worst

Bumble users talked about how it can be easier to make friends on the app than in-person because they know everyone on Bumble BFF is open to making friends. However, all Bumble users talked about feeling unsure if a particular person was interested in being their friend. This anxiety seemed to be particularly strong at two times: first, when looking at a list of matches and thinking about starting concersations with new people. All Bumble users reported feeling like they never got messages from new matches, and that they always had to message first. Second, when thinking about moving beyond messaging to an in-person interaction. Again, ll Bumble users said that if they didn't invite matches to meet up, then nothing would ever happen because conversations would slowly fizzle out.

1:1 hangouts are essential to strengthen friendships

Several of the participants who were introverted or reported having social anxiety said that ideally they would build friendships exclusively through 1:1s. Other participants wanted to meet in a group setting first, and then meet 1:1. Some preferred to meet 1:1 first, and then meet in group settings, before eventually meeting 1:1 again in order to go from being a friend to a close friend. No matter which camp, people eventually circled back to 1:1s, and several participants articulated that is because people are more vulnerable and grow closer in 1:1 settings; you don't become best friends in a group setting.

Little concensus about group hangouts

Some participants were interested in group meetups, half of them said they would only be interested if they picked all the attendees, and the other half said they would never be interested in planning a group meetup. Preferences were very split on whether they would be willing to show up and meet strangers, whether seeing profiles ahead of time would be enough, or if they wanted to text in a group or text 1:1 before meeting up in person. This, combined with how split people were on what kinds of activities would help them make friends, makes it seem like planning group activities would be a niche market among Bumble users.

Recommendations

Finding someone with similar interests: All participants talked about looking for friends who have similar interests - hobbies they do, causes they care about, etc - and how they would like to see that information on people's profiles. Within a month of conducting these interviews, Bumble began A/B testing and then released an app update that allows users to showcase their stage of life (eg, new in town, wedding planning, going back to school, working parent, sober life) and their interests (eg, religion, social activism, sports, music). While I'm sure some people wish there were more categories and subcategories, I think this will be a huge part in helping Bumble users find people with matching hobbies!

Finding someone to click with: Several participants articulated that profiles help them weed out people who they wouldn't be friends with (eg, people they would disagree with too much politically or religiously), but don't help them find someone who they'd would be friends with (eg, someone they "just click with"). The dating app Hinge has a feature where users are asked a) if you went on a date and b) if your date is the type of person you'd like to see again. A feature like that might help Bumble have more accurate data to give people better matches.

Help people move beyond texting: All Bumble users talked about feeling like they had to instigate with every person they met on Bumble, and most participants talked about worrying if other people were really interested in being their friend. Bumble might try to help get the ball rolling by having a chat bot asking one or both participants if they're interested in meeting up in person, or jumping in with suggestions to both people in the chat. "Hi Naomi, we noticed you've been chatting with Sarah for about two weeks. Are you interested in meeting them?" If the user selected "Yes", Bumble could suggest a few kinds of activities.

Group meetups: It seems like there is some interest in group meetup, but there was no consensus on how many people would work, how people should connect before a meetup, and how to help people feel comfortable. At this time, Bumble should focus on making 1:1 meetups feel more natural.

Limitations

Keep moderator's guides short: Because I have no affiliation with Bumble and only my personal experiences as background information, I went into this project blind. My moderator's guide had way more questions than it should have for a 60 minute interview, and I did run out of time for the last scenario or two on a few interviews. I easily could have done a full 60 minute interview on any of the three research questions. It's usually better to do quality research than to get a large quantity of feedback, but this time it seemed worthwhile because I needed to get so much background information.

Have a budget for research: I wish I could have talked to more people who had never used Bumble, but I talked to the people who were willing to give me an hour of their time. Several participants were people that I met on Bumble, so I asked them to not talk about meeting me in the interview in order to not get biased data - but ideally, I wouldn't be interviewing my friends about friendship.

Try other research methods: I think talking to people before going to a 1:1 or group activity, observing the activity, and then talking to them individually afterwards would have been much more insightful than just talking to people about the idea of meeting strangers. People are notoriously bad at predicting what they will think or do. Another option could have been a diary study - while there are drawbacks, I would have been able to see how people's impressions and opinions before an activity, see their report, and what they thought afterwards. Ideally, I would love to talk to both people before and after they met to see both sides.