clock menu more-arrow no yes mobile
The Marriage Pact, an algorithm that removes endless swiping and choice from the online dating experience, went viral at Stanford two years ago.
Javier Zarracina/Vox

The dating algorithm that gives you just one match

The Marriage Pact is designed to help college students find their perfect “backup plan.”

The Highlight by Vox logo

Siena Streiber, an English major at Stanford University, wasn’t looking for a husband. But waiting at the cafe, she felt nervous nonetheless. “I remember thinking, at least we’re meeting for coffee and not some fancy dinner,” she said. What had started as a joke — a campus-wide quiz that promised to tell her which Stanford classmate she should marry — had quickly turned into something more. Now there was a person sitting down across from her, and she felt both excited and anxious.

The quiz that had brought them together was part of a multi-year study called the Marriage Pact, created by two Stanford students. Using economic theory and cutting-edge computer science, the Marriage Pact is designed to match people up in stable partnerships.

As Streiber and her date chatted, “It became immediately clear to me why we were a 100 percent match,” she said. They found out they’d both grown up in Los Angeles, had attended nearby high schools, and eventually wanted to work in entertainment. They even had a similar sense of humor.

“It was the excitement of getting paired with a stranger but the possibility of not getting paired with a stranger,” she mused. “I didn’t have to filter myself at all.” Coffee turned into lunch, and the pair decided to skip their afternoon classes to hang out. It almost seemed too good to be true.

In 2000, psychologists Sheena Iyengar and Mark Lepper wrote a paper on the paradox of choice — the concept that having too many options can lead to decision paralysis. Seventeen years later, two Stanford classmates, Sophia Sterling-Angus and Liam McGregor, landed on a similar concept while taking an economics class on market design. They’d seen how overwhelming choice impacted their classmates’ love lives and felt certain it led to “worse outcomes.”

“Tinder’s huge innovation was that they eliminated rejection, but they introduced massive search costs,” McGregor explained. “People increase their bar because there’s this artificial belief of endless options.”

Sterling-Angus, who was an economics major, and McGregor, who studied computer science, had an idea: What if, rather than presenting people with a limitless array of attractive photos, they radically shrank the dating pool? What if they gave people one match based on core values, rather than many matches based on interests (which can change) or physical attraction (which can fade)?

“There are a lot of superficial things that people prioritize in short-term relationships that kind of work against their search for ‘the one,’” McGregor said. “As you turn that dial and look at five-month, five-year, or five-decade relationships, what matters really, really changes. If you’re spending 50 years with someone, I think you get past their height.”

The pair quickly realized that selling long-term partnership to college students wouldn’t work. So they focused instead on matching people with their perfect “backup plan” — the person they could marry later on if they didn’t meet anyone else.

Remember the Friends episode where Rachel makes Ross promise her that if neither of them are married by the time they’re 40, they’ll settle down and marry each other? That’s what McGregor and Sterling-Angus were after — a sort of romantic safety net that prioritized stability over initial attraction. And while “marriage pacts” have probably long been informally invoked, they’d never been powered by an algorithm.

What started as Sterling-Angus and McGregor’s minor class project quickly became a viral phenomenon on campus. They’ve run the experiment two years in a row, and last year, 7,600 students participated: 4,600 at Stanford, or just over half the undergraduate population, and 3,000 at Oxford, which the creators chose as a second location because Sterling-Angus had studied abroad there.

“There were videos on Snapchat of people freaking out in their freshman dorms, just screaming,” Sterling-Angus said. “Oh, my god, people were running down the halls trying to find their matches,” added McGregor.

Next year the study will be in its third year, and McGregor and Sterling-Angus tentatively plan to launch it at a few more schools including Dartmouth, Princeton, and the University of Southern California. But it’s unclear if the project can scale beyond the bubble of elite college campuses, or if the algorithm, now operating among college students, contains the magic key to a stable marriage.

The idea was hatched during an economics class on market design and matching algorithms in fall 2017. “It was the beginning of the quarter, so we were feeling pretty ambitious,” Sterling-Angus said with a laugh. “We were like, ‘We have so much time, let’s do this.’” While the rest of the students dutifully fulfilled the class requirement of writing a single paper about an algorithm, Sterling-Angus and McGregor decided to design an entire study, hoping to solve one of life’s most complex problems.

The idea was to match people not based solely on similarities (unless that’s what a participant values in a relationship), but on complex compatibility questions. Each person would fill out a detailed survey, and the algorithm would compare their responses to everyone else’s, using a learned compatibility model to assign a “compatibility score.” It then made the best one-to-one pairings possible — giving each person the best match it could — while also doing the same for everyone else.

McGregor and Sterling-Angus read through academic journals and talked to experts to design a survey that could test core companionship values. It had questions like: How much should your future kids get as an allowance? Do you like kinky sex? Do you think you’re smarter than most other people at Stanford? Would you keep a gun in the house?

Then they sent it to every undergraduate at their school. “Listen,” their email read. “Finding a life partner is probably not a priority right now. You hope things will manifest naturally. But years from now, you may realize that most viable boos are already hitched. At that point, it’s less about finding ‘the one’ and more about finding ‘the last one left.’ Take our quiz, and find your marriage pact match here.”

They hoped for 100 responses. Within an hour, they had 1,000. The next day they had 2,500. When they closed the survey a few days later, they had 4,100. “We were really floored,” Sterling-Angus said.

At around 11 pm the following Monday, they sent out the results. Instantly, the campus went wild. Resident assistants texted them saying the freshmen dorms were in chaos, and the Stanford memes Facebook page — where students share campus-specific humor — was awash in Marriage Pact content.


Streiber, the English major who would go on to meet her match for coffee and discover how much they had in common, remembers filling out the survey with friends. Amused at this “very Stanford way” of solving the school’s perpetually “odd dating culture,” she wrote a tongue-in-cheek poem about the experience:


In the following weeks, McGregor and Sterling-Angus began to hear more about the matches. “People were saying they were matched with their exes, with their best friend’s boyfriend,” Sterling-Angus recalled. “Siblings matched, and everyone else was horrified but we were ecstatic because we’re like, ‘It works.’”

A few people started dating their matches, but that was almost beside the point. The flaws they’d seen the first year could be easily fixed — there were simple ways to make sure no one matched with their siblings — but for now, their proof of concept had worked. It already felt like a win.

The Marriage Pact’s focus on core values echoes that of older dating sites like OkCupid, which gives users a list of potential mates with compatibility scores based on a questionnaire. But OkCupid still runs into the issue of presenting people with seemingly infinite options. Meanwhile, newer apps like Tinder and Hinge, which emphasize profile photos, were built for endless swiping, compounding the paradox of choice.

These dating apps are “competing to keep you swiping for as long as possible,” summarized Tristan Harris, the co-founder and director of the Center for Humane Technology. “They get you addicted to getting attention ... and try to turn your social life into Las Vegas.”

Some apps have tried to rectify this problem by restricting the supply of potential matches and encouraging people to meet in person as soon as possible. In June, Bumble, an app designed around women making the first move, opened a wine bar in SoHo called Bumble Brew. Two years earlier, they’d opened a pop-up restaurant called Hive. “The lines were out the door,” according to a report by Bloomberg.

While the League, a dating app for people with “high standards,” has no such storefront, it purposely tries to limit the dating supply. “Instead of endless swiping, users receive between three and seven matches a day, and we aim to make them quality potentials that could be your future soulmate,” wrote Amanda Bradford, founder and CEO, in an email. “It’s impossible to predict chemistry and nothing beats meeting in person, so all of the features that we are working on are designed to get people to meet in person as quickly as possible rather than judge a book by its cover,” she added.

But McGregor and Sterling-Angus are doubtful these strategies will ultimately result in sustainable relationships. The Marriage Pact, they argue, doesn’t prioritize user engagement. Its purpose is to actually find you someone you could partner with for life.

“What’s a successful outcome on the apps, a phone number exchange?” McGregor asked. “No, a successful outcome is staying on the app,” Sterling-Angus corrected.

“It’s a hookup and then return.”

Today, the dating app market is an estimated $3 billion industry, and more than half of all single people in the US have tried a dating app at some point. Online dating isn’t going anywhere, however frustrating people find it to be. If the Marriage Pact is able to scale beyond college campuses, it could provide a welcome alternative to the typical swiping experience.

McGregor and Sterling-Angus aren’t sure when that will happen, but they are already well into designing their next study. “We’re not gonna make this good, we’re gonna make this really good,” McGregor said.

Next year, they want to bring the Marriage Pact to more schools, including state schools on the East Coast, using a network of friends and colleagues to determine where it would be most successful. “For now, we operate in pre-filtered communities,” Sterling-Angus said. She knows this is part of their success, since “people are fairly like-minded and have a strong sense of affiliation” at universities.

And after that? Sterling-Angus and McGregor told me they eventually hope to launch the Marriage Pact in “other communities that still have a strong sense of identity” but declined to comment on specifics, saying they hadn’t “finalized that internally” yet. When asked if they could see doing this full time, they said, yes, if the study ever expanded to become a company.

If and when that happens, Sterling-Angus and McGregor will have to contend with the ramifications of taking money from investors who have their own ideas about what “success” means for online dating, and the exponentially messier problem of matching people up in a dating pool outside the college elite.

They’ll also have to answer the question that looms over the Marriage Pact: In the long term, can the algorithm actually lead to happy, lasting relationships? Does it work?

After Streiber graduated from Stanford, she moved back to LA to pursue acting full time. But she hasn’t forgotten about her Marriage Pact match. She told me that after their first coffee date, she followed up with him to try to hang out, but they never seemed to find a time. “I kept going back to our first conversation and being like oh, it went so well, what happened, what changed? But one thing I realized is that for as perfect a match as we were on paper, that doesn’t always translate into real life.”

When we spoke on the phone, Streiber was on her way to an improv show. “It’s summer now, he’s back, and I might be seeing him tonight, in a weird twist of events,” she said.

She texted me the next day: “Just wanted to let you know that the guy never ended up showing last night! Classic.”

The Highlight

Everything old is new again

The Highlight

The billionaire’s guide to self-help

The Highlight

Crypto: New. Fraud: Old.

View all stories in The Highlight