Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
Algorithmic dating, like online dating in general, has overpromised and underdelivered. Yes, that is the uninformed opinion of a centrist dad who has never tried online dating — but it is not merely the uninformed opinion of a centrist dad who has never tried online dating. The market has reached the same conclusion: despite a background of frothy valuations for all things tech, share prices of the dating companies Bumble and Match Group (the owner of Tinder) have fallen precipitously in the past few years. User numbers are flagging too.
This reckoning has been brewing for a very long time. Scientific matchmaking dates back at least to the 1920s, when Science and Invention magazine explained the use of pulsometers (“electrical sphygmographs”) and a body odour test (put the object of your affection in a glass capsule to which a hosepipe is attached, and sniff away). One test even had a researcher suddenly fire a pistol into the air to see how the loving couple reacted to stressful situations — if both of them showed signs of panic, this was alleged to be a bad omen for the chances of marital harmony. But a more direct precursor of today’s dating apps is Operation Match, a 1960s effort by enterprising Harvard students who were scrounging some time on a punch-card computer.
Operation Match claimed to rely on a 75-point questionnaire to make the perfect pairings, but the truth was simpler. “The first thing we did was to make sure they were in the same area,” co-founder Jeff Tarr later told a documentary-maker. “Mostly girls wanted to go out with boys who were the same age or older, their height or taller, the same religion. So after we had these cuts, then we just kind of randomly matched them.”
So much for scientific matchmaking, but there are worse ways to find love. While Grindr and Tinder are slicker and more immediate, they seem to work in much the same way, eschewing an algorithmic match in favour of what so many web ads describe as “hot singles in your area now”. (If I am the only one seeing these ads, please never tell me this.)
Perhaps that is sensible. There is obvious appeal in the idea that from a sphygmograph to a deep-learning system, “science” will find your perfect match, but it is not hard to see the shallowness in the promise of an algorithmic pairing. Mathematician Hannah Fry — author of The Mathematics of Love — put her finger on the problem: “You don’t really know what you want.”
We can write down the list of qualities we might want in a partner, but some of them are circular (“I would be attracted to someone attractive”), some of them seem important but may not be (such as a shared taste in books or music) and some defy description. A list of categories in a computer database might feel scientific, but we should hardly be surprised if our affections are governed by a very different subconscious checklist.
Also, people lie. Researchers have discovered that the typical user of online dating websites is richer, slimmer, blonder and sexier than the rest of the population — at least, according to their own profiles. Lake Wobegon is real, and accessible on a dating app near you.
A further complaint is simply that dating apps induce people to spend a lot of time anxiously scrolling and not nearly enough time going out and having fun. This is 21st-century life anyway, but the gap between what the app promises and what the app delivers is particularly stark when the app is offering to help you find love — or, at the very least, some kind of intimacy with another human.
Looming over all this is a broader social question: are dating apps bad for society? The worry here is not prudish but more of a parallel with social media. We worry about Twitter, YouTube and TikTok not only because they distract and distress us but because they may be contributing to a polarised society in which everyone lives in their own information bubble.
A recent working paper written by three economists, Yujung Hwang, Aureo de Paula and Fanzhu Yang, tries to shed light on the question of whether dating apps polarise us. As with social media there are forces pushing in both directions: some dating apps allow people to filter who they see by categories including race, religion and education. These filters might contribute to greater polarisation, where people date only people of the same race and education level. On the other hand, dating apps make it easier to skim through a large number of possible matches, just as a social media account presents a vast range of different hot takes. So perhaps dating apps encourage more mixing across ethnic or educational lines?
My bet was on increased polarisation. Faced with a broader choice of people to connect with, we often use that choice to seek out people just like us. Consider a study of college friendships conducted by three psychologists, Angela Bahns, Kate Pickett and Christian Crandall. They compared the friendships that students formed at small college campuses with those at the much-larger University of Kansas. The smaller campuses offered less diversity overall, yet students at small colleges were more likely to have a diverse group of friends — diversity here referring to all sorts of differences in opinion, background and behaviour. Given more choice, people sought out their ideological twins.
Despite this, Hwang and colleagues found that the impact of online dating was — to my surprise — to enable people to date and eventually to marry people of a different ethnic background. Why? The answer is simple: Tinder, the most popular app, does not offer filters beyond basics such as location and sexual orientation. Instead, users are thrown into an undifferentiated pool of dating prospects and have to figure it out.
There is a strange parallel to social media here: although social media are filtered and highly polarised, they are also chaotic and scattershot. Filter bubble notwithstanding, you are more likely to encounter opposing viewpoints on Twitter than reading your favourite tabloid newspaper.
Chaotic and scattershot doesn’t sound like a recipe for dating success any more than it sounds like a healthy news diet. But in both situations, there is a case to be made for casting the net widely and in strange waters — as long as we can occasionally move past the shallows into something deeper.
That deeper experience might be a good book, a serious hobby or a long-term romance. The only problem: there’s no revenue in any of that.
Find out about our latest stories first — follow FT Weekend Magazine on X and FT Weekend on Instagram
