Remember what dating was like in Junior High? I do. Full disclosure: I didn't do a lot of it. Nevertheless, I have keen memories of the extensive negotiations that went on before people actually started "going out." Conversations went something like this: "Do you like him? He likes you." Self-appointed matchmakers served as go-betweens, ensuring that there was mutual interest before anyone ever publicly announced anything -- a primitive but effective form of double opt-in.
Sean Rad, CEO of the enormously popular dating app Tinder, uses the ancient art of the double opt-in to help broker love matches. I had a chance to interview Sean Rad after his talk at GigaOm Roadmap, and got to hear more about how Rad and his team have leveraged a fundamental understanding of human behavior to fuel Tinder.
Carrie Yury: I loved hearing you talk about Tinder. The app is very simple and fun, but it's informed by some really deep thinking about human motivations. Can you tell me a little bit about how you use human experience to influence the design?
Sean Rad: We built an experience that we wanted and a flow that emulates interactions in the real world. What we're striving to do with Tinder is understand what are the sort of social dynamics, physical dynamics, the flows of where you start with a request or desire to make a new relationship and how that progresses to you actually meeting that person, talking to that person and getting to know them. We emulate the flow of the app against that model.
Carrie Yury: Does it matter whether your purpose as a user is to hook-up or to find your husband or wife?
Sean Rad: I think it emulates whatever you want in the real world. So if you are young and you don't want to be in a serious relationship, you are going to look for that on Tinder, or if you are older and desire something more serious you will look for that.
I don't think when people go out, they really have an outcome in mind. Particularly with millennials, they sort of just go with the flow and one thing leads to another. A short-term relationship might lead to a marriage. You'll never know. A short-term desire might lead to a long-term outcome.
Carrie Yury: The front end of Tinder is really fun and engaging, especially because it's so simple. What is happening on the back end? How do you make sense of all those swipes?
Sean Rad: We look at your behavior and we optimize who we show you based on who you are saying yes or no to. There are a lot of signals that we take into consideration. If you say no to somebody there are a lot of things about that person that we know -- whether you had common friends with them, who the common friends were, how old that person is, on and on, what their interests are. We take all that into consideration when serving better recommendations in the future.
Also, when you match with somebody, we look at the depth of the conversations you are having with your various matches. You might have a deeper conversation with one person of a certain characteristic or another person of another different characteristic.
Carrie Yury: I really like what you had to say about asking as little as possible of users before they get started with the app. You described it as having a kind of negative "ripple effect" that could prevent users from engaging. Is that something you talk about a lot?
Sean Rad: I haven't talked about it publicly, but internally we talk about it all the time. It's one of our philosophies. A 1 percent ask could have a 20 percent impact on a user experience. We're just very careful, not only to not ask the wrong things because we don't want to disrupt the user experience, but also asking certain things creates a certain context with the user that we might not want to create.
For example, there are some apps where you sign up and they ask you questions like, "Do you belong to this group or that group? Are you this or are you that?" And by making these comparisons and categorizing the individual, you are automatically alienating them to some degree. You're making them think about something that they otherwise might not have wanted to think about coming into that experience.
It goes back to the analogy of, when I walk into a room and go to a party, I'm not forced to answer questions about what my political beliefs are and what my religious beliefs are. I am just there mingling and meeting people and one thing leads to another and maybe that comes up in a conversation organically, but I don't feel out a survey when I answer the door saying that, "I'm Jewish and and I am conservative, on and on." I just sort of am who I am and I walk in and I have relationships and discover about other people and share about myself as I progress.
Carrie Yury: Your app is so situated in what people need, what they want, and how they would do things outside of the digital context. Is that a mind frame you've always had?
Sean Rad: Always. I often tell the team, "Throw away any academic approach to user experience. Throw away all of that. It's all bullshit." It all comes down to what do people want to do, how do they want to do it, and how do we create a frictionless experience to allow them to do that? How does the interface, the functionality, create a story around what the desired action is you are trying to create for people? Whether we are coming up with a feature or improving something, we always look at, first and foremost, how does that relate to some desire or some thought process that the user has.
Carrie Yury is Director of Research at BeyondCurious.