The Marriage Pact: Looking for Love Through An Algorithm

By Ipshita Faldu

Tulane Marriage Pact Promo- @tulanemarriagepact on Instagram

About 2 weeks ago, 2300 students at Tulane University opened their emails to find the name of their ideal match. Doesn’t everyone wish they had a backup plan? One person, one optimal match, who you could marry if you don’t find anyone else. Marriage pact provides a romantic safeguard that will prioritizes long term stability over the initial attraction that is usually a huge part of modern online dating. By removing the huge array of attractive pictures and pretentious captions, Marriage Pact aims to minimise your dating pool. They give people one match, based on core values rather than similar interests and physical attraction. In the era of socially distant dating, Marriage Pact offered an opportunity that was never seen before. It was a chance to meet your campus soulmate, while doing the bare minimum.

        The Marriage Pact was originally started in 2017 by two undergraduates at Stanford University. While the rest of their classmates wrote a one page paper about algorithms as part of an Economics class, these students decided to create an algorithm and base an entire study that could potentially solve one of the most complex problems of life. Each person would fill out a detailed survey, and the algorithm would compare their responses to everyone else’s, using a compatibility model to assign a ‘score.’ It then made the best one-to-one pairings possible .The combination of studies in psychology, market design and computer science allowed them to design a survey that could test compatibility based on core values.

       The survey began with demographic questions like class year, gender identity, sexual orientation, political stance, race and religion before getting into questions carefully tailored to understand the participants’ outlook on life. Questions ranged anywhere from ‘Is there a place for revenge when someone wrongs you’ and ‘would you ever own a gun’ to ‘how would your friends describe you’. The Marriage Pact works with school partners to tailor the process to each school. While the majority of the questions remain the same across colleges, a few key questions are based on the culture of the campus.

      Kelsey, a sophomore at Tulane University was a part of the team that helped bring Marriage Pact to Tulane. She described her role as one of the school ambassadors for the event.‘I just saw it on TikTok. I never thought I would actually get involved in it. She dmed the Marriage Pact on Instagram never expecting a response back. Three weeks later, she was a part of a team that was ready to officially launch Marriage Pact at Tulane. The process required one week of orientation to understand and learn how to work around the algorithm.The rest of the time was spent making graphics and doing marketing leading up to the launch. “We had to make sure to make the questions personalised to Tulane”. While the basic questions remained the same, the team went through a database of questions provided by the Marriage Pact to determine what would suit the audience at Tulane the best. According to Kelsey, the party culture especially had to be taken into account for a school like Tulane. Questions involved asking the participants’ their opinion on alcohol, hard drugs and popular bars.

       A number of marketing and advertising strategies were used to spread the word. One thing that got really popular among the students was ‘Hot Takes’. This was an email they sent out to generate interest while they were running the algorithm; the hot takes were responses you gave that a very small population of the survey takers answered similarly. This was a teaser for the student as they could see how the algorithm actually worked. For example, one of the “Hot takes’ questions stated: “There is a place for revenge when someone has wronged me.” You were supposed to place yourself somewhere between 1 (turn the other cheek) and 7 (plotting rn)”. Jordan, a junior at Tulane who took part in the Marriage Pact thought this was a great idea.‘I answered “1”, which only 7% of respondents said too. It just seemed like a very specific and unusual question, but I suppose it could tell you a lot about how a partner keeps or discards bad feelings.’Giving results and showing a part of how the algorithm works was an effective way to involve students. Students saw the algorithm in action and were even more motivated to take part. 

Marriage Pact official website-

       However, Jordan was not very pleased when she took her marriage pact questionnaire. When asked if she thought the questions were a good reflection of who she was as a person and what she was looking for, Jordan said, “Not at all. This is partly because they said they were matching on both similarities and differences, but I got the sense it was based mostly on similarity of responses. It seemed like people got random incompatible matches because most relationships form from proximity and social circles.” While the questions in the Marriage Pact were carefully selected to ensure a long term, stable match, it does seem like they focused more on deeper life questions that students don’t really think about. “None of us figure out the things that were on the survey until we are deeper into knowing someone, and at that point, many other factors could have turned us onto or off of the person before finding out their deeper preferences.But ironically, I think the fact that they didn’t ask too many questions about small preferences like sense of humor and cultural interests made many of the matches unsatisfactory at the superficial level.”

       While the Marriage Pact emphasized core values in order to move away from superficial, physical attributes, it was ironic how it ended up predominantly coming around that. Students understood the primary goal of the questionnaire but most participants still ended up making a judgement based on their matches’ social media appearances.When asked about her match, Jordan said, “I would not have considered him in a million years and still wouldn’t, largely for immediate superficial qualities I noted from looking at his social media after matching. He seemed a little bit like a douchey Tulane boy I would avoid.”This is one aspect about the Marriage Pact methodology that students seem to disagree with. “I do think there could be a screening period where they give people a few close matches and have them rank their attractiveness to match people with a partner they’re also attracted to. However, I think Gen Z solves this problem themselves by checking out their match on their own online”, Jordan said.

      “People weren’t happy”, Kelsey said. Students at Tulane saw the Marriage Pact as something that would help them find the person they were going to stay with for the rest of their lives.  For the team at Tulane, it was only a fun way to connect to people, expand social circles and find people who had similar values. After the results of the survey came out, The Tulane Marriage Pact instagram received multiple dms asking for a ‘new match’ or potentially ‘taking the test again’. Another reason why Marriage Pact wasn’t as much of a success at Tulane was the unequal female to male ratio. About 900 straight girls who filled out the questionnaire did not receive a match. Marriage Pact instead gave these girls a ‘friend match’. This was another girl who shared the same values as the participant according to the algorithm. According to Kelsey, a huge reason for this was the all-girls launch team at Tulane. The graphics and the marketing leading up to the launch were designed from a female perspective. “Next time I wish we emphasized the idea that this isn’t a dating service”. Kelsey said when talking about their plans next year. The Marriage Pact will be coming to Tulane again next year. They hope to target a wider audience. “We got dms from people saying they had fomo”. There were multiple groups they didn’t reach out to and a lot of people didn’t know this was happening. Maybe this would help with the number of male participants in the pact. 

       It has only been 2 weeks since results came out. While some students lost hope when they saw their matches, some are still meeting their matches and potentially building a relationship with them. Only time will tell if the algorithm actually works. In the meantime, Marriage Pact is constantly improving, changing their questions and their algorithm to get better at finding students their one true match. What Marriage pact did succeed in doing was building excitement and providing students with a fun story to tell.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s