Dan Ariely is the James B Duke Professor of Psychology and Behavioral Economics at Duke University. He is the founder of The Center for Advanced Hindsight and co-founder of BEworks,[…]
Online dating could be a crucial tool for single people, but with the sites we have now you’ll likely spend six hours searching for every date you go on.
Question: What did you find when you started looking into the world rnof online dating?
rn
rnDan Ariely: I became interested in online dating because one of rnthe people who were sitting in an office next to me was incredibly rnmiserable, and he was an assistant professor; he just moved to the rnuniversity where I was at; he was spending long hours; he was not rnfinding anybody to date; he was, couldn’t date students at the rnuniversity, he was a professor; he didn’t have time to go outside. You rnknow, we were not particularly a social bunch, you know, he was rnbasically stuck. And online dating was a very promising way to think rnabout this solution for a marketplace that wasn’t working very well, andrn he tried online dating and he was just failing miserably, rncontinuously. So that kind of piqued my curiosity about it. And then Irn started looking at online dating.
rn
rnSo I start looking by registering myself and looking at other people andrn then I said, let me ask some of my friends to enroll. So I didn’t ask rnthem to really enroll, I just took their profile sheets and asked rnpeople, "Could you fill those out but without your name?" And I took rnpeople that I liked more and I liked less, and I took their profile and Irn tried to figure out could I tell the difference? You know, now, rnimagine you did this. Imagine you went to 50 people you really like andrn 50 people you only like so-so, and you asked all of them to fill this rnprofile, then you took this 100 profiles and you tried to sort them out rninto piles. Turns out we’re terrible at this! Right? So this is kind rnof an initial observation that something is going wrong in this, in thisrn market.
rn
rnAnd then went a step further, did some studies with online daters about rnhow much they enjoyed it and what they were getting from it, until the rnfinal stage, we, I figured out, I thought I knew what was going on, rnwhich is that online dating sites assume that people are easy to rndescribe on searchable attributes. They think that we’re like digital rncameras, that you can describe somebody by their height and weight and rnpolitical affiliation and so on. But it turns out people are much more rnlike wine. That when you taste the wine, you could describe it, but rnit’s not a very useful description. But you know if you like it or rndon’t. And it’s the complexity and the completeness of the experience rnthat tells you if you like a person or not. And this breaking into rnattributes turns out not to be very informative.
rn
rnSo on the last stage of this process, we created a different Web site. rnAnd that different website allowed people to experience other people rnwithout all of these attributes. And we show that this is actually much rnbetter and would lead to much more, much higher probability of going on arn second, on a real date afterward. So it kind of goes from an rnobservation to a little study, to a bit more details and then finally rnproposing some kind of solution of something that I think would actuallyrn work better.
rn
rnSo the site basically looks at real dates—and think about what real rndates are. They’re not about sitting in the room and interviewing each rnother about questions; they’re often about experiencing something rntogether in the real world. And I think it’s because if you and I went rnout, and we went somewhere, I would look at how you react to the outsidern world. What music you like, what you don’t like, what kind of picturesrn you like, what kind of images, how do you react to other people, what rndo you do in the restaurant? And through all these kind of non-explicitrn aspects, I will learn something about you and I would feel that I’m rnlearning something about you. And the online system we created was veryrn much like that. It was about you came up and you got a little avatar, arn square or a triangle, some color, and you went into a virtual space in rnwhich you could explore it. And you could see lots of stuff, there werern pictures and images and there were words and there movies and there rnwere bands, there was all kinds of stuff, and you could go and when you rncame to another little avatar, you could start chatting. And you would rnchat about something, it wasn’t about interviewing when you went to rnschool and what’s your religion; it was about talking about something rnelse and it turns out it gave people much more information about each rnother, and they were much more likely to want to meet each other for a rnfirst date and for a second date.
rn
rnQuestion: What's wrong with the experience of online dating?
rn
rnDan Ariely: I think that online dating is an incredibly rnunsatisfying experience. In fact, when we do surveys to understand whatrn people do, the basic trade off is for each six hours of searching for rnpeople and emailing them, you get one cup of coffee. And it's not as ifrn people enjoy online dating, it's not as if they have fun searching rnpeople and writing blurbs for them. I mean, imagine that you basically rnhad to drive six hours, three hours each way to have coffee with rnsomebody, and, you know, coffee usually ends up with just coffee. It's rnan incredibly unsatisfying experience. So I think it's a really bad, rnit's a really bad system.
rn
rnOn top of that, there's another thing, is which, imagine I gave you thisrn search criteria, which I asked you to search by height and weight and rnincome and all of those things: you're going to use it. That's what I rngive you to search, you're going to use it. There's a million people rnout there, you want to limit them to 3,000, that's what we're going to, rnthat's what you're going to use. And because of that, I think actually rnpeople become much more superficial than we think they are. So here's rnan example. It turns out, women really care about men's height. I’m rn5’9”, if I wanted to be as attractive as somebody who’s 5’10”, right, rnanother inch? I would have to make about $35-40,000 more a year. rnThat’s a lot of money for one inch. At the same time, it turns out thatrn men care a lot about women’s BMI’s. In fact, they want women to be rnslightly anorexic, at like 18-1/2. And you look at women’s rnattractiveness, it goes really up at low BMI and really drops below rnthat.
rn
rnNow, people online look incredibly superficial. They look at hair colorrn and they look at height and they look at income, and that’s basically rnit... and attractiveness, of course. And you can ask, is it because rnthat’s all people care about or is that because that’s what the system rnis giving them to search for. And I think it’s because of combination, rnright? Sure, we are superficial, we do care about attractiveness and rnheight and income and these are features for us, but I think they’re rnexaggerated by the way the system is created.
rn
rnImagine you were looking for something else, imagine you were looking rnfor digital cameras, and imagine that I only allow you to search on rnmegapixels and f-stop for the lens, right? These things would become rnincredibly important, right? And if I drop some things from the search,rn they would become as if they’re not important or much less important. rnSo I think part of the problem is that the systems don’t give us the rnright information that we need. And because of this, I think the rnexperience of online dating is generally unsatisfying. I mean, think rnabout it: how many millions of people are participating in this activityrn and marriage rates has not increased, divorce has not decreased. I rnmean, not really much has happened because of that. And at the same rntime, I think it’s incredibly important, right? The dating market is rnperhaps the only market that we moved from a centralized market to a rndecentralized market. You know, we used to have a yenta, your parents rnused to tell you what to do, all this is gone, now you have to fend for rnyourself. On top of that, we move a lot, right? You go to one place rnfor undergrad, then you go to grad school, then you move to another cityrn for a job, two years later you move again. You have no time to create arn social network. We work long hours, so it’s really a system where we rndon’t have time to find people for ourselves. It’s taboo to date people rnat the work place, the social networks are weaker in the physical rnworld. We move all the time and we don’t have a yenta or parents to rntell us what to do.
rn
rnSo online dating are incredibly important, it could be central and rncrucial and we need to create them because it’s really a miserable rnsituation for most single people. At the same time, the ones that we rnhave created, and they all look the same basically, they’re no real rndifferences between them, the ones that we are creating are just not rnthat useful.
Recorded on June 1, 2010
Interviewed by David Hirschman
rn
rnDan Ariely: I became interested in online dating because one of rnthe people who were sitting in an office next to me was incredibly rnmiserable, and he was an assistant professor; he just moved to the rnuniversity where I was at; he was spending long hours; he was not rnfinding anybody to date; he was, couldn’t date students at the rnuniversity, he was a professor; he didn’t have time to go outside. You rnknow, we were not particularly a social bunch, you know, he was rnbasically stuck. And online dating was a very promising way to think rnabout this solution for a marketplace that wasn’t working very well, andrn he tried online dating and he was just failing miserably, rncontinuously. So that kind of piqued my curiosity about it. And then Irn started looking at online dating.
rn
rnSo I start looking by registering myself and looking at other people andrn then I said, let me ask some of my friends to enroll. So I didn’t ask rnthem to really enroll, I just took their profile sheets and asked rnpeople, "Could you fill those out but without your name?" And I took rnpeople that I liked more and I liked less, and I took their profile and Irn tried to figure out could I tell the difference? You know, now, rnimagine you did this. Imagine you went to 50 people you really like andrn 50 people you only like so-so, and you asked all of them to fill this rnprofile, then you took this 100 profiles and you tried to sort them out rninto piles. Turns out we’re terrible at this! Right? So this is kind rnof an initial observation that something is going wrong in this, in thisrn market.
rn
rnAnd then went a step further, did some studies with online daters about rnhow much they enjoyed it and what they were getting from it, until the rnfinal stage, we, I figured out, I thought I knew what was going on, rnwhich is that online dating sites assume that people are easy to rndescribe on searchable attributes. They think that we’re like digital rncameras, that you can describe somebody by their height and weight and rnpolitical affiliation and so on. But it turns out people are much more rnlike wine. That when you taste the wine, you could describe it, but rnit’s not a very useful description. But you know if you like it or rndon’t. And it’s the complexity and the completeness of the experience rnthat tells you if you like a person or not. And this breaking into rnattributes turns out not to be very informative.
rn
rnSo on the last stage of this process, we created a different Web site. rnAnd that different website allowed people to experience other people rnwithout all of these attributes. And we show that this is actually much rnbetter and would lead to much more, much higher probability of going on arn second, on a real date afterward. So it kind of goes from an rnobservation to a little study, to a bit more details and then finally rnproposing some kind of solution of something that I think would actuallyrn work better.
rn
rnSo the site basically looks at real dates—and think about what real rndates are. They’re not about sitting in the room and interviewing each rnother about questions; they’re often about experiencing something rntogether in the real world. And I think it’s because if you and I went rnout, and we went somewhere, I would look at how you react to the outsidern world. What music you like, what you don’t like, what kind of picturesrn you like, what kind of images, how do you react to other people, what rndo you do in the restaurant? And through all these kind of non-explicitrn aspects, I will learn something about you and I would feel that I’m rnlearning something about you. And the online system we created was veryrn much like that. It was about you came up and you got a little avatar, arn square or a triangle, some color, and you went into a virtual space in rnwhich you could explore it. And you could see lots of stuff, there werern pictures and images and there were words and there movies and there rnwere bands, there was all kinds of stuff, and you could go and when you rncame to another little avatar, you could start chatting. And you would rnchat about something, it wasn’t about interviewing when you went to rnschool and what’s your religion; it was about talking about something rnelse and it turns out it gave people much more information about each rnother, and they were much more likely to want to meet each other for a rnfirst date and for a second date.
rn
rnQuestion: What's wrong with the experience of online dating?
rn
rnDan Ariely: I think that online dating is an incredibly rnunsatisfying experience. In fact, when we do surveys to understand whatrn people do, the basic trade off is for each six hours of searching for rnpeople and emailing them, you get one cup of coffee. And it's not as ifrn people enjoy online dating, it's not as if they have fun searching rnpeople and writing blurbs for them. I mean, imagine that you basically rnhad to drive six hours, three hours each way to have coffee with rnsomebody, and, you know, coffee usually ends up with just coffee. It's rnan incredibly unsatisfying experience. So I think it's a really bad, rnit's a really bad system.
rn
rnOn top of that, there's another thing, is which, imagine I gave you thisrn search criteria, which I asked you to search by height and weight and rnincome and all of those things: you're going to use it. That's what I rngive you to search, you're going to use it. There's a million people rnout there, you want to limit them to 3,000, that's what we're going to, rnthat's what you're going to use. And because of that, I think actually rnpeople become much more superficial than we think they are. So here's rnan example. It turns out, women really care about men's height. I’m rn5’9”, if I wanted to be as attractive as somebody who’s 5’10”, right, rnanother inch? I would have to make about $35-40,000 more a year. rnThat’s a lot of money for one inch. At the same time, it turns out thatrn men care a lot about women’s BMI’s. In fact, they want women to be rnslightly anorexic, at like 18-1/2. And you look at women’s rnattractiveness, it goes really up at low BMI and really drops below rnthat.
rn
rnNow, people online look incredibly superficial. They look at hair colorrn and they look at height and they look at income, and that’s basically rnit... and attractiveness, of course. And you can ask, is it because rnthat’s all people care about or is that because that’s what the system rnis giving them to search for. And I think it’s because of combination, rnright? Sure, we are superficial, we do care about attractiveness and rnheight and income and these are features for us, but I think they’re rnexaggerated by the way the system is created.
rn
rnImagine you were looking for something else, imagine you were looking rnfor digital cameras, and imagine that I only allow you to search on rnmegapixels and f-stop for the lens, right? These things would become rnincredibly important, right? And if I drop some things from the search,rn they would become as if they’re not important or much less important. rnSo I think part of the problem is that the systems don’t give us the rnright information that we need. And because of this, I think the rnexperience of online dating is generally unsatisfying. I mean, think rnabout it: how many millions of people are participating in this activityrn and marriage rates has not increased, divorce has not decreased. I rnmean, not really much has happened because of that. And at the same rntime, I think it’s incredibly important, right? The dating market is rnperhaps the only market that we moved from a centralized market to a rndecentralized market. You know, we used to have a yenta, your parents rnused to tell you what to do, all this is gone, now you have to fend for rnyourself. On top of that, we move a lot, right? You go to one place rnfor undergrad, then you go to grad school, then you move to another cityrn for a job, two years later you move again. You have no time to create arn social network. We work long hours, so it’s really a system where we rndon’t have time to find people for ourselves. It’s taboo to date people rnat the work place, the social networks are weaker in the physical rnworld. We move all the time and we don’t have a yenta or parents to rntell us what to do.
rn
rnSo online dating are incredibly important, it could be central and rncrucial and we need to create them because it’s really a miserable rnsituation for most single people. At the same time, the ones that we rnhave created, and they all look the same basically, they’re no real rndifferences between them, the ones that we are creating are just not rnthat useful.
Recorded on June 1, 2010
Interviewed by David Hirschman
