Funny thing is that when dating apps used to be "browse profiles, send message", nobody had to debate inner workings. Users could self select people of interest, send them a message directly, and know there'd be a very good chance that their profile/message would at least be seen.<p>Issue today is that apps control visibility, both in terms of profiles and likes. I've tried Hinge a few times. Did like a 4 month initial stint that had me consistently matching/meeting very attractive women (ones in the standouts section), tried it again a year later under a paid plan and had one mediocre match in 2 weeks (same photos and profile).<p>Skimmed the article so maybe this was addressed, but there's dark patterns happening on these apps, or faulty algos, or both.
Interesting data in the article though nothing unexpected for people who follow this space. Some notable points:<p>> The other thing that interests you is the like ratio, or the openness, among 100 profiles that the user sees, how many of them does he like? (The median for men is 26% and for women is 4%.)<p>>The like ratio of a girl is almost independent of the profiles she sees. For example, if a girl has a like ratio of 5% and you remove 50% of the profiles, even if you remove only the profiles she will not like, her like ratio will still be 5% (you can do that by removing very unattractive people for a guy that is very attractive, for example). It is funny to observe, but it seems like a girl has internal reasoning on a dating app, and they know they can only like x% of profiles whatever she sees (of course, it doesn't work if you show only ugly people).<p>And lastly:<p>>Whats interesting is that the more attractive the guys were ranked by girls the more they were looking for something not serious.
Retention is a terrible metric for a dating app. The "perfect" dating app would have a 0% retention rate since the first person you meet would be your "ideal partner".<p>It's tantamount to measuring a hospital's performance by it's retention rate.
> To me, if you are a guy on a dating app and your pictures are not taken by a professional photographer then you are losing your time, and if you are paying you are also throwing your money.<p>Don't do this.<p>You need good pictures that convey attractiveness (looks, as well as personality). Using professional photos conveys neediness & a level of desperation hidden under a shell of an ego the shot tries to portray. So you end up relying on looks with a handicap. A good looking person doesnt need professional shots to show that.<p>Sure, if you currently have mirror selfies, professional shots are better. Otherwise - if you are not a model who has magazine-published shots you're including in your profile, then don't go use or pay for professional shots. Figure out how to take canned shots on your own or pay a photographer for canned real shots (nothing highly edited).
My best dating apps lately have been Trader Joe's and Ducky's Car Wash.<p>I wear aloha shirts every day, and nearly every time I go to TJ's, someone asks me where they can find a particular item. It may be a guy or a gal, but I am always happy to help a neighbor find what they need.<p>That is not the reason I wear aloha shirts. I just love these shirts! Every spring I get the Cooke Street shirts at Costco, one of each new pattern.<p>One time at the Menlo Park Trader Joe's I was talking with the guy restocking the freezer section. He said, "Nice aloha shirt! I bet people sometimes think you work here."<p>Sure enough, a minute later a young lady walked up to me and asked if we had organic bread. I walked her over to the bread section and pointed out the organic breads.<p>Later I caught up with the freezer guy again and told him "you were right!"<p>Ducky's, for those unfamiliar, is a car wash with several locations on the SF Peninsula. Even if you just get an exterior wash, after you go through the tunnel they hand dry your car.<p>There is a waiting area outside with a dozen chairs, and it takes 5-10 minutes before your car is ready.<p>And you never know who you might run into there!<p>The key to this, of course, is to be outgoing and friendly, and open to surprises.
We built a dating app and saw these same metrics.<p>Our "twist" was that anyone could be a matchmaker, pairing other people (in addition to our recommendation engine).<p>An interesting thing happened: lot's of people just like using dating apps in a voyeuristic way, with no intention of dating.<p>Tinder eventually launched this feature, but it fell flat.<p>I still think there's merit to human-based recs over "AI" matchmakers.
The qualities that women find attractive in men cannot be captured in a static profile on a website ... plus most of the men in that site are there because they don't have those qualities.<p>Let's just say that if you can open your mouth enough to say hi to a woman without hesitation, you are completely wasting your time on dating sites.
There is a truly fascinating talk by a data scientist who "hacked" the "algorithm" of a dating site and became the most matched person in californina.<p>Highly recommended.<p><a href="https://www.youtube.com/watch?v=mJod9kRYyao" rel="nofollow">https://www.youtube.com/watch?v=mJod9kRYyao</a>
Has anyone noticed that the histograms have some spikes and valleys that are probably artifacts?<p>My guess is that there's some rounding or floating point shannanigans going on.<p>Yes this is what I have to contribute to the conversation, I cannot speak to the dating dynamics as I am an unexpert on that subject
Dating apps are nothing more than window shopping. The lowest effort, most shallow, unlimited options buffet of meat sacks to choose from.<p>When you have unlimited options, and a potentially better one is just a swipe away, people become disposable. Fuck or not fuck. Hot or not. Bang or pass. Where in the past someone might tolerate an insignificant quirk someone has, nowadays such a thing is often all that it takes to ghost and keep looking.<p>The users that get a lot of matches often put in zero effort into talking. They expect you to do all the conversation, and often act uninterested, if you ever get a reply after the match at all. Matching and never getting a reply after contact is more common than not, regardless of how much thought and time is put into the first message.<p>Profiles often read more like a list of demands, a quota that must be reached before you'll even be considered worthy. This is true for all the genders.<p>"Make me laugh" is a demand that I've seen on countless profiles. Entertain me monkey, make me laugh. So many "Make me laugh"s that I've started to wonder if most of the profiles I see are actually bot/scammer accounts, because so many of them have slight variations of the same exact profile bios. AI is only going to make that shit show worse, and I'm sure many people are already trying to wine and dine an AI profile and have no clue.<p>Let's also not forget about the clear "looking for a greencard" mail order bride profiles, and the weird as fuck profiles where each picture has a child as the focus of every picture, very intentionally so, in a way that I can't quantify but when you run across them, it just feels off, weird, saying something without saying something.<p>Dating apps are all exactly the same anymore. Same fake profiles. Same gamification of attraction that you have to pay to win or have to grind every day against those who do.<p>Dating apps are not there to match people. That would put them out of business. They are there to extract funds from wallets while keeping you hopeful that your perfect 10 is one more premium teir up the chain, one more swipe away, one more billing cycle in the future.
Equal parts fascinating and dystopian.<p>I'm even more convinced now that online dating has reached a local optima, but eventually someone is going to find a solution that is less shallow and predatory and blow it out of the water.
> More than 50% of men just never receive a like, and never means maybe 2 or 3 likes in the lifespan of several weeks<p>Half the user base are patsies is basically the fundamental design.
He ends with a very interesting proposition to "fix" dating apps:<p>> To me, the next revolution is really concepts that will make you meet other people without having much information about them. As a user, you will trust the algorithm to match you with the right people. And these concepts will be paying only<p>> So to summarize, a concept where users pay and commit that they will meet people without knowing them before. So yes, it will take a few dates to really find someone that you like, but so is going every day in dating apps and meeting people that you don't like either.<p>This is very close to the Dutch app Breeze: <a href="https://breeze.social/" rel="nofollow">https://breeze.social/</a>. There's no chatting in the app. It's focused on meeting people as soon as possible. People pre-pay for dates (covers drinks) and the app partners with venues to check on the couples (they know their names). People who cancel dates get a badge on their profile saying that they have canceled. Ghosters get banned.
My anecdotal user-end data-science-ish story about dating apps:<p>A few years back I was single and on Hinge a fair amount. If you used Hinge back then, you'll remember some key differences between the platform and other dating apps: 1) when you "like"'d someone, you'd have to comment on a specific part of their profile (a photo, a prompt answer, etc), 2) these likes showed up in their inbox, independent of whether they liked you or not (as in, you didn't have to like each other mutually; the other end decided whether to reply or ignore after delivery), and 3) there was limit per day, you could like/message 8 profiles per day, no more. On average, swiping through my 8 per day, I'd generally get 1-2 new replies, which turned roughly into 3-4 first dates per month.<p>One of the key elements is that the inbox was time-ordered: the most recent like you received was at the top. There was discussion on the Hinge subreddit about how girls would typically only click through the top few items in their inbox daily, and if you were lower down, you were doomed to drown under the mountain of new message they're getting on top. So I figured I'd solve for "what is the optimal time of day to be blasting out my likes to ensure I end up higher in the inbox?"<p>You can probably see where this is going: I requested a GDPR data export, which happened to have all my conversations, time-stamped. Crunching through in Python there was something in the data I didn't really expect.. a disproportionate number of first-replies (replies to my initial like/message, that is) were around the 2-3pm bucket. Not what I would've expected (don't these people work?) but fair enough, I started doing all my swiping in those hours instead of in the evening as I usually did.<p>And it worked. Good god did it work. I consistently started getting replies to 70-80% of my initial messages (from the ~10% before). I was drowning in conversations to the point where I wouldn't swipe at all for days for fear of yet another conversation to manage. Within a few months I ended up meeting my current girlfriend and haven't been back on since, but it was surprising how well something simple like time-of-day affected my reply rate.
> The likelihood to like and exchange inside homosexual groups is much higher than in heterosexual ones.<p>> The like ratio of a girl is almost independent of the profiles she sees. For example, if a girl has a like ratio of 5% and you remove 50% of the profiles, even if you remove only the profiles she will not like, her like ratio will still be 5%<p>These two statements sound like they would be at odds. It seems either the first statement is incorrect or women on dating apps are more choosy when it comes to men only. I’d be curious how the stats play out on lesbian dating apps
I paid for a lifetime membership to Bumble premium a couple years ago, but as I noticed matches/likes declining for no apparent reason, I made a 2nd account with different credentials, paid for a week of premium, and had a bunch of matches (exact same profile as my other account).<p>From my anecdotal experience, they have some algorithm that leads to diminishing returns in order to keep extracting money for boosts or whatever.
This is such a detailed article but it's giving me weird vibes.<p>For instance there are all these drops to near-zero in the histograms at .28, .46, .56 for no clear reason, and the article doesn't even consider that noteworthy.<p>The "Men Like ratio (y) vs ratio (x)" has an inexplicable wall around .33 which I could only explain with some sort of product limitation maybe? But I really wish it was explained what artifacts the product introduces.
> Recommendation of profiles that you may like is a solved technical challenge at Tinder level and at mostly any dating app today.<p>It’s hard to take the rest of the article seriously after reading this!
I worked in this space briefly. This is a really good article in line with the conclusion of the OK Cupid! team which were scrubbed from their web site but are in this book:<p><a href="https://en.wikipedia.org/wiki/Dataclysm" rel="nofollow">https://en.wikipedia.org/wiki/Dataclysm</a><p>If you're a guy and you want to depress yourself, find a female friend who is signed to a dating app and see all the likes and messages she gets. When I was in this space, an attractive woman was getting on the order of 1000 messages <i>a day</i>. 95% of those messages were one-worders, "hi", "wassup". Basically a total tsunami of garbage.<p>LLMs are going to change this space enormously as they are going to act as your agent to talk to the other person's agent. The world of online dating is going to be horrible.<p>I already know people who have dozens of verified female dating app profiles staffed by LLMs and overseas operators that they use to trick men into clicking links to generate revenue.<p>Enshittification all round.
To me, it reveals that women are not really interested in dating.<p>Women go on app when they feel a bit sad, when they need to feel a bit better.<p>People want to stay single. Introversion is on the rise.
I'm pretty sure the first four graphs alone already prove that dating apps don't work.<p>Dating apps are supposed to match people, but desire to match up is very lopsided towards one gender, with the other gender having very little desire to match up.<p>Having unrealistic expectations is one thing. Being the monkey paw that fulfills those wishes is on a wholly different level.
Retention is at odds with the goals of the users of dating apps. I wonder if there’s a more aligned system that charges for premium features when there’s inactivity. Or you can only use the app’s premium features again after some period of inactivity if you pay<p>Inactivity serving as a potential signal that you found a match, or you’re no longer in the market. Although the incentive system would have to ensure the app isn’t aligned to make the experience so annoying you stop using it either
The data here falls in line with the infamous OKcupid study (which got cancelled and taken down because men and women are identical, donchaknow?)<p>The takeaway is that humans best date by meeting people in person through mutual acquaintances.<p>Without the forced direct social interaction, women are only interested in the top 10% of guys, and guys are just aimlessly running at anything that moves regardless of their actual interest (i.e. liking and seeking sex from women they have no real interest in dating). Guys end up with no likes and no dates and women end up with mountains of disingenuous likes and dates with disingenuous men.
> most guys don't open the profiles of a girl, and girls open much more often but more for discarding a user than for looking for more information<p>this is a factor of why it is a loosing strategy to do dinner, drinks and coffee dates<p>first the user experience is an arduous interview that is reliant on a “spark” that leaves too much to chance<p>secondly, the ask is for somewhere you can hear each other, “get to know each other”, and have one foot out the door easily to leave<p>on the woman’s side…. its the same as this article is saying: for disqualifying, not qualifying.<p>for the guys side, whatever your goal is, would be accomplished by disregarding the woman’s preference. in reality, she likely did go on a more elaborate and rambunctious date, generated endorphins (the spark), got more intimate than her risk models dictate, and <i>then</i> realized she didn't know the guy.<p>if you want any of that to happen, be like the other guy<p>it is optimal to be like the 5% guy that is closing with women. if you cannot replicate that on dating apps, you may be able to with other in person approaches and date ideas
A great article that elicited a lot of thoughts.<p>(1) I used to make those kind of non-informative scatter plots with xvgr when I was a grad student, this package does a great job for those kind of cases<p><a href="https://seaborn.pydata.org/generated/seaborn.relplot.html" rel="nofollow">https://seaborn.pydata.org/generated/seaborn.relplot.html</a><p>even if you don't use it you can copy its patterns to make designs that work<p>(2) An obvious commercial offering for guys is a photography package. About 20 years ago I went to the biggest photog in my town and my publisher paid $100 for a headshot that was just a junior photog in the studio. If you were a bride you would get premium hair and makeup to go with your photography, even if you were appearing on TV you would probably get a little hair and makeup help.<p>(3) With the right choice architecture you could control things such as "the percentage of people that you like" or "the number of likes that you receive". For instance if you were going just on looks it would be easy to show people a stack of 10 photos and have them sort them in attractiveness; you could also show pairs of profiles and pick an ELO for each one. If you look at it as a relative ranking process you can peel off whatever percentage off the top that you want.<p>An obvious objection is that given such a choice the "hot" people will be the only ones that get chosen but a counter to that is that you can put an upper limit on how many "likes" somebody gets by not showing them to people.<p>This contradicts some things he says later on about things that help the apps retain people, but from the viewpoint of making an app that "works", girls who are looking for commitment really aren't benefiting from seeing profiles from hot guys who get a lot of attention and provide nothing but casual sex.
Turns out even highly motivated professionals don't understand that correlation is not causality:<p>> For girls, it is the number of likes sent; the number of likes received has no impact on retention, maybe a little bit but less than 1%. The number of likes sent has a huge impact; a user that liked no profile in her 100 first scrolls has a d30 of 12%, and 19% for girls that like 10 profiles and 16% for girls that liked 5 profiles. The d1 retention is almost 100% correlated to a girl sending 5 likes to active guys in the first 24 hours (the real thing is to get a match, but it is easy to get a match when a girl sends 5 likes). So to have the perfect d1 retention for girls, the only thing you should focus on is to get them to send 5 likes. And you have about 100 scrolls to do so.
> More than 50% of men just never receive a like, and never means maybe 2 or 3 likes in the lifespan of several weeks<p>As someone in that more than 50%, it’s very annoying to constantly get told to get on the apps to meet women. I’m surrounded by men in the top 20% because I’m affluent, well educated, and spend a lot of time at the gym. Sadly, I’m just around these people and wasn’t born into the same kind of family. I’m an outsider. I was born poor and ugly. I’ve solved the poor thing but being ugly is incurable. I’m going to Beverley hills next week and getting more surgery to try to alleviate the ugliness but it’s pathetic what a man in his mid-30’s has to do now to even get a single like back on his profile.<p>Women don’t need men anymore in the developed world. Men are luxury goods and women are completely happy to live without. A man isn’t needed but merely wanted and only wanted if he fits a very particular set of criteria.
The advice to boys/girls felt very real. And also, objectively not equal.<p>Girl: lower your standards.<p>Boy: pay a photographer to look nicer.<p>This double disadvantages women as I see it: Their standards are held to be unrealistic (downpoint) and they have to incur the dissatisfaction of beleiving it, and acting on it to lower their expectations. AND Boys are expected to glam up and project a demonstrably less real state of themselves, to get over this bar. So women have to accept lies, and tolerate reality.<p>(63, in a longterm 40y relationship, not using the apps and not judging individuals here)
> You would need to be a pro at user interviews to really get interesting feedback. (Well maybe this is the norm in B2C but at the end of the day user interviews were of no help).<p>This shouldn’t be surprising. Interviewing humans is a skill. Doing so in a product context, and learning useful things from it, is not easy.<p>I hope they don’t approach other things this way. “You’d need to be a professional plumber to stop water leaking out of this. Maybe that’s the norm but at the end of the day plumbing was no help.”
Thanks for this detailed post.
While I think most are aware of the broad situation, it's impressive to read it all laid out like that. And whenever the author tried to make a statement about guys, the gist of it was "we don't know, it doesn't matter, boys will stick around". Made me chuckle.
There are a lot of dating apps out there with very different audiences. I think the mistake made here is generalizing the conclusions for all these apps. Women date differently than men. And not always with men. Or only with certain types of men.
I'm not going to go into details as I don't want to create a throwaway account for HN, but I can attribute a lot of people's feelings in dating apps to a few things. I got an email from Bumble a few years ago that said I was in the top x percent of people swiped on.<p>If you try to brute force stats your way to dating apps, you will fail.... to some extent.<p>A lot of this comes down to looks that you can control, and looks that you cannot control. Some people are born better looking than others and when you spend less than a second filtering people, the first factor you use is looks. That said, not everyone is looking for the same qualities so ymmv, but better looking people find dating apps much easier.<p>Throwing money at apps works. I'm not going to go into details because my opinion is not based on anything other than my opinion, but I found that the more I spent on the apps, the more dates I would get.<p>Modern dating when compared to traditional dating offline is not even the same thing. Ghosting and talking romantically to multiple people is normal. You can't let yourself get emotionally attached to anyone until you actually know them or expect anything from them.<p>I've heard horror stories from both men and women from online dating, and I've only had great exeriences from it. Some people find me attractive, and at the time I was very active and fit, so I usually got past the swipe test. I'm honest with myself and ok with my flaws. I'm also comfortable in social situations which helps me talk to new people.<p>I think crunching the numbers in this style only looks at a binary 'reality' of dating apps and not what you can do to help yourself and other factors that can lead you to what you ultimately want from partnership, or relationships or physical comfort or whatever else lead you to online dating.
> There is also face detection that is now pretty advanced and can detect the face of celebrities, people on social media.<p>Which naturally results in celebrities and models getting blocked from signing up periodically.
It's funny because you can tell they barely discussed with actual girls:<p>> they are used to receiving a lot of likes, so you have to show them they get liked<p>They hate that, most girls registering on Tinder get 1000 likes in a week.<p>> Dating fatigue is bullshit<p>See last point<p>> Having a profile doesn't impact the app experience<p>That's what changed the game, nobody bothers on Tinder because it's not mandatory whereas it works very well on Hinge.
Dating apps are scams. The developer is out to make money, which happens through subscription renewals. Profile visibility means nothing. Messages mean nothing. You can send message after message after message, but unless you have spent enough money or are onenof the lucky ones that the system allows through, your profile is not shown to anyone but bots.<p>Want to test this? Remove your image from your profiles on social media and remove your last name (i use my middle name as my surname on facebook). The key is to remove your profile image. Set your privacy to maximum, so other images of you cannot be searched. Try talking with someone. With no way to profile you, the chat bots used by the dating apps cannot have a "conversation." You will never be matched, conversations will hit walls and go in circles or you get ghosted because the chat bot has no data to use.<p>After your first renewal - after you have spent more money - then you may actually get to talk to a real person.<p>No. Never again.
> Can Dating apps be fixed?<p>> What I meant by "fixed", is an app where:<p>> It is possible for someone to reach out to anyone and get an answer, and discussion can be interesting from the start (at least as much as in real life)<p>> Your looks, how you behave, the tone of your voice,... reflect what you look like in real life<p>> It is easier to meet new people than attending local events<p>> I don't think it can<p>Well, as long as the metrics app builders are optimizing for are "retention" and "monetization", like this post obsesses over, and the people building the apps continue to refer to women as "girls", like they've never had a relationship with an adult woman, then I agree, dating apps are going to continue their process of enshittification.
What I hate about Tinder is that it has become the only way to find woman for sex and for relationship, even if with miserable probabilities. Even in my tiny non English speaking country. The man has to pay for the app. Women ending up on Tinder loathe how despicable the app is ("looking for a pearl in this cesspool", "don't believe I ended up here" etc.) while somehow don't notice that they make this place this way.
> Among girls : 10% where lesbian, 84% were hetero and 12 % were interested in both genders<p>> Among guys: 7% homo, 92% hetero and almost none were interested in both genders<p>That's insane.