Amy Webb: Can You Use Algorithms To Find Love? Amy Webb was having no luck with online dating, so she started treating the world of online dating as data — effectively hacking her way to finding a spouse.

Amy Webb: Can You Use Algorithms To Find Love?

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


It's the TED Radio Hour from NPR. I'm Guy Raz. And our show today - ideas around how we love. So love is instinctive, right. It's buried deep inside the most primitive part of our brains.

But the way it works - why we're drawn to certain people and not others - it's still not entirely understood. So this hour, TED speakers who are all exploring love and not just romantic love, but the kind of love you might feel for your dad or your brother or best friend. So our first story is a romantic, love story with help from math, data and algorithms as told on the TED stage.


AMY WEBB: My name is Amy Webb. And a few years ago, I found myself at the end of yet another fantastic relationship that came burning down in a spectacular fashion. And I thought what's wrong with me? I don't understand why this keeps happening. So I asked everybody in my life what they thought. I turned to my grandmother, who always had plenty of advice, and she said stop being so picky.

You've got to date around. And most importantly, true love will find you when you least expect it. In short, I was trying to figure out, well, what's the probability of my finding Mr. Right? Well, at the time I was living in the city of Philadelphia. And it's a big city. And I figured, you know, in this entire place, there are lots of possibilities. Population of Philadelphia has 1.5 million people, figure about half of that are men.

So that takes the number down to 750,000. I'm looking for a guy between the ages of 30 and 36, which was only four percent of the population. So now I'm dealing with a possibility of 30,000 men. I was looking for somebody who was Jewish 'cause that's what I am and it was important to me, so only 2.3 percent of the population. I figure I'm attracted to maybe 1 out of 10 of those men. And there was, you know, no way I was going to deal with somebody who was an avid golfer. So that basically meant there were 35 men for me that I could possibly date in the entire of city of Philadelphia.


WEBB: So if I have two possible strategies at this point - I'm sort of figuring out. One, I can take my grandmother's advice and sort of least expect my way into maybe bumping into the 1 out of 35 possible men in the entire 1.5 million-person city of Philadelphia or I could try online dating. Now, I like the idea of online dating because it's predicated on an algorithm. And that's really just a simple way of saying I've got a problem.

I'm going to use some data, run it through a system and get to a solution. So in my case, I thought will data and an algorithm lead me to my Prince Charming? So I decided to sign on. Now, the biggest problem is that I hate filling out questionnaires of any kind. And I certainly don't like questionnaires that are like Cosmo quizzes. So I just copied and pasted from my resume.


WEBB: So in the descriptive part up top I said that I was an award-winning journalist and a future thinker. When I was asked about fun activities and, like, my ideal date, I said monetization and fluency in Japanese. I talked a lot about JavaScript.


RAZ: Fluent in Japanese and JavaScript. That is - that's hot.

WEBB: That's super sexy.

RAZ: That's sexy.

WEBB: Yeah.

RAZ: Yeah.

WEBB: Yeah, maybe that wasn't the best way for me to introduce myself. But, you know, the crazy thing is that even though I had foolishly copied and pasted from my resume, it didn't stop the dating services from matching me with other people. And it certainly didn't stop those people from asking me out on dates.

RAZ: So how did they go?

WEBB: Yeah. I had some dates that were pretty rough. I was being set up with very, very Orthodox rabbis, which was, like, a no-go from the get-go, people who were super interested in sports. There was being stuck with the check. There was a another guy who was diminutive and ordered a lot of Long Island Iced Teas.

And we were out doing karaoke on our first date, and he ran up on stage and sang a bunch of songs and then dedicated them to his girlfriend. And I was like, I have no idea who you are. I just met you, like, 20 minutes ago. I am not your girlfriend.

RAZ: Now, the thing you should know about Amy Webb is that she crunches numbers for a living. She analyzes data that helps big companies make more accurate predictions. And so she started to wonder, what if love isn't so mysterious at all? What if, instead of analyzing data for her clients, why not do it for herself? And what if, by doing that, she could game the system?

WEBB: You know, in any other case, I would do market research. Why wouldn't I do market research, you know, on myself? So dating websites are sort of predicated on some pretty basic, not very exciting math. And in order to make things work, there has to be a limited number of choices, a limited number of variables. So it's a lot easier to parse do you like cats or dogs than it is to parse something like chemistry, right.


WEBB: Knowing that there was superficial data that was being used to match me up with other people, I decided instead to ask my own questions. What was every single possible thing that I could think of that I was looking for in a mate?

So I was looking for compatibility in terms of work ethic, religion.


WEBB: So I started writing.

You know, musical tastes...


WEBB: ...And writing.

I wanted somebody who had a certain attitude towards money...


WEBB: ...And writing.

I was looking for somebody who was going to be 20 pounds heavier than I was at all times.


WEBB: Somebody who was going to be totally OK with forcing our child to start taking piano lessons at age 3.

It was a pretty exhaustive list.


WEBB: And at the end, I had amassed 72 different data points, which, to be fair, is a lot. So what I did was I went through and I prioritized that list. I broke it into a top tier and a second tier of points. And I ranked everything starting at 100 and going all the way down to 91. So once I had all this done, I then built a scoring system.

What I wanted to do was to sort of mathematically calculate whether or not I thought the guy that I found online would be a match with me. I figured there would be a minimum of 700 points before I would agree to email somebody or respond to an email message. For 900 points, I'd agree to go out on a date. And I wouldn't even consider any kind of relationship before somebody had crossed the 1,500-point threshold.

RAZ: And this is, like, unbelievable. You are like Alan Turing, like, cracking the enigma code. You've cracked the online dating code.

WEBB: Well, I cracked if for myself. And I think that's what this comes right down to. A lot of people are - they either go into relationships not really knowing what they want and they change or they've settled. And when you make your list, when you really think about who it is that's going to make you happy in the long term and what you're going to need, that should be the time that you make the most detailed list of your entire life.

I know people who have a handful of things they're looking for in a mate, but who have grocery lists that are three pages long. You are grocery shopping for a soulmate. There isn't a lot of science behind cracking the code. It's about figuring out what you need to make you happy and then going out and getting it. You know, in my case, I didn't want to go out on 50 dates. I wanted to go out on one date with the right person and be done.


WEBB: Well, as it turns out, this worked pretty well. So I go back online now, I found JewishDoc57, who's incredibly good looking, incredibly well-spoken. He had walked along the Great Wall. He likes to travel as long as it doesn't involve a cruise ship, right. And I thought I've done it. I've cracked the code. I have just found the Jewish Prince Charming...


WEBB: ...Of my family's dreams. There was only one problem - he didn't like me back. And I guess the one variable that I haven't considered is the competition. Who are all of the other women on these dating sites? I found SmileyGirl1978.


WEBB: She said she was a fun girl who is happy and outgoing. She listed her job as teacher. She said she is silly, nice and friendly. She likes to make people laugh a lot. At this moment, I knew, clicking after profile, after profile, after profile that looked like this that I needed to do some market research. So I created 10 fake, male profiles. Now before I lose all of you...


WEBB: ...All right, understand that I did this strictly to gather data about everybody else in the system. I didn't carry on crazy catfish-style relationships with anybody. I really was just scraping their data. But I didn't want everybody's data. I only wanted data on the women who were going to be attracted to the type of man that I really, really wanted to marry. And mainly what I was looking at was two different data sets.

So I was looking at qualitative data - so what was the humor, the tone, the voice, the communication style that these women shared in common - and also quantitative data - so what was the average length of their profile, what - how much time was spent between messages? I wanted to figure out how to maximize my own profile online. And as it turns out, I did a really good job. I was the most popular person online.


WEBB: And as it turns out, lots and lots of men wanted to date me. Well, not too long after that, I found this guy. And he said that he was culturally Jewish. He talked in detail about travel. He looked and talked exactly like what I wanted. And immediately, he scored 850 points. It was enough for a date.

Three weeks later, we met up in person for what turned out to be a 14-hour-long conversation that went from coffee shop to restaurant to another coffee shop to another restaurant. Well, a year and a half after that, we were non-cruiseship traveling through Petra, Jordan when he got down on his knee and proposed. A year after that, we were married. And about a year and a half after that, our daughter, Petra, was born.


RAZ: That's incredible. It's like a movie. I mean, it's amazing that that happened, that all that happened.

WEBB: It is. So afterwards, I eventually did show him the list. So fourth date in I had said, listen, I got to tell you something.

RAZ: Yeah.

WEBB: And I took the list out, and I said here's how we came to be together. And he thought that it was great. One of the things that was on the list was I was looking for somebody who would appreciate the beauty of a well-crafted spreadsheet.

RAZ: Yeah. That's totally - that's exactly the right way to go.

WEBB: Well, and it was, and he did.

RAZ: Wow. I mean, so if technology is, like, changing, you know, the way we find love, right. And if the algorithms can be gamed - I don't know - couldn't it, like, lead to the perfect person, like, the person you are meant to be with forever?

WEBB: I think technology is a really useful tool to bring people together. But at the end of the day, it's up to us. Technology has made a lot of things in life much more efficient, much easier. Love is something that takes work.

And it takes work even if you found your soulmate, your 1,500-point man or woman, the person that you are looking for who is the perfect person for you. You both still have to put in some effort. And technology can't solve for that critical element of any relationship. For love to endure, it takes human capital. It takes sweat equity, understanding, and it takes people.

RAZ: Amy Webb, she told her story in a memoir. It's called "Data: A Love Story." Her full talk is at In a moment, the science behind who you love. Our show today, how we love. I'm Guy Raz, and you're listening to the TED Radio Hour from NPR.

Copyright © 2014 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.