Hacker News Reader

Comment by pixelpoet | Hacker News Reader

Okcupid used to be really good, then they got bought out and turned into Tinder. Now it's just a wasteland out there, especially for guys.

View on HN

OkCupid was really well thought out for making matches for relationships. The double connection nature of the questions that allowed not just answering questions but also what answers you'd accept and how strongly you felt about those answers was great. It also did a nice job of sussing out people's true personalities. The more questions they answered the more difficult it was to hide their true selves. One of my favorite examples of this was a question about why birds don't get harmed when they land on power lines. The question could be used to gauge a person's technical knowledge but the answer "They do but they express it poorly" was a signal of a sense of humor. Lots of questions were variations on each other but written differently which is another way of getting to a person's true core.

OkCupid was full of weirdos, though. Source: I was one of them.

Tinder hit on to a psychological hack that allowed normal people to use it. It's something to do with plausible deniability. Setting up a real dating profile signals that you know what you want but implies that you are lonely/unhappy. Tinder gives that vibe of "I'm just swiping, look at all these losers, I'm just having fun". So normies went on it too.

Why doesn't someone just make a new site that works like the old OkCupid and isn't owned by Match Group?

The OKCupid model doesn't have the same short-term profits and/or VC returns as desperate people shelling out for dating app subscriptions, so nobody will fund it.

How much does it actually cost? You set up a website. You put ads for chocolates and flowers on the site so you can recover your expenses. It's not like you have to build a state of the art fab. It's basically a messaging app, which is the sort of thing individuals have built as a side project.

Building it is easy, moderating it is hard

Moderating it is spam filtering. You limit signups to IP addresses in the the country of your target market and then ban signups from any IP address that tries to send an excessively large number of messages or has an excessively high block rate.

At that point you're down to real users who are jerks, the solution to which is the block button and a message sorting algorithm that takes into account how many times it's been used against someone.

The why don't you do it. It'll be worth billions right.

Now tell me all the reasons you won't do it.

No, it won't be worth billions, that's kind of the point. It should be possible to do it efficiently so that the operating costs are low and it doesn't need to make billions in order to be sustainable and outcompete the incumbents.

As to why I don't do it, maybe I will -- but I wouldn't object if someone beat me to it. Because it still requires some combination of time and money, the amount of which can be feasible without being zero, and it's quite likely that of all the people in the world, someone who isn't me is in a better position to pull it off before I do.

> You put ads for chocolates and flowers on the site so you can recover your expenses

This isn't going to be bring in anywhere near the amount of money you're thinking it will.

It's not supposed to make a lot of money. It's supposed to cover expenses so you don't go out of business if you do get a hundred million users.

> just make a new site that works like the old OkCupid

Is such a thing even possible in today's world where attention spans are measured in seconds thanks to a decade of social media and endless pursuit of "engagement"?

People are all burned out on "engagement" and hankering for something legitimate.

Network effects. Your business could run the worlds most perfect dating app but it doesnt matter if it only has 10,000 users globally.

Dating apps were easier to bootstrap back in the day before the whole market was swallowed up.

They all start with less than 5000 users. Everyone is constantly complaining about the incumbents. Go to where they're complaining and tell them you've done it properly. Since you actually have, they tell others.

That's not how network effects work. Your complainers will go back to their subreddit and complain that your dating app is empty and all the profiles they're matched up with are 500km away.

The only way to have a fighting chance is to start in a single metropolis (eg New York), and try to get everyone to coordinate trying at once. Since the demographics of people complaining about dating sites online is a bit small (and selects for the kinds of clients you don't want), you've got to advertise more broadly, eg with ads on Youtube or in the metro. That gets expensive.

You don't actually need everyone to sign up at once. Once someone signs up you send them emails when they get a match. When there aren't as many people they don't get as many emails, but now they're on the site, which creates more matches for people who join tomorrow.

People need some way to find out about it, but not everything has to be corporate. Wikipedia has this page which ranks near the top for search queries like "list of online dating sites":

https://en.wikipedia.org/wiki/Comparison_of_online_dating_se...

You're an online dating service, so you get added to pages like that where people end up when they're looking to choose an online dating service. And then your site compares well against the other ones that are screwing everyone -- look how few of the heterosexual dating apps have free messaging. So people sign up and give it a try.

You make sure your site is listed in places like that where people go to find dating sites, and people find it. And the more people find it, the more useful it gets, because that is how network effects work. At which point people start recommending it and you get even more users.

Another possibility is allowing people to sign up for waiting lists, and as soon as you have enough people in certain locations, you let them in. There's lots of possibilities of launching this correctly.

Sure, there's a lot of clever ways to get an audience as a dating site, and I've seen many sites with clever marketing tactics. The success rate is still abysmal.

I doubt this is actually a problem. People seem to yearn for places to express themselves in that aren't run by parasites

One problem that a new site/app would face is that, to be useful, it would need a geographical density of users. If you have 100k people signed up, but they're spread out across north America or Europe or wherever, then very few of those people will ever meet irl. So at best you have a messaging site for lonely people, but more likely just an unsustainable business. Achieving the necessary user density needs scale and advertising budget.

Something highly local might work in a big enough city like Paris/London/NYC.

This is the reason the Childfree dating site I'm on is entirely useless. There are literally dozens of us! All spread out geographically.

There was a really nice app that would literally create a curated dating app profile for you and free appointments with a dating coach etc. I ran into a while back, unfortunately all of the users were so far away it was completely useless lol

I believe they patented their matching system, so i guess you'd run into legal battles, and they would have deeper pockets.

Edit: Match and others definitely did get patents, so I may be confusing them, but I guess no matter who, you'd get sued by someone.

OkCupid is from 2004. Patents last 20 years. This can't be an excuse for much longer, and software patents are bullshit anyway.

Why should it matter whether you implement your algorithm with atoms that are arranged into transistors or ones that are arranged otherwise?

Why would any algorithm be patentable, as opposed to actual inventions, like a new battery chemistry or a new kind of light bulb?

Because it became 5x as hard to do this well in 2023. The legal and community landscape changed immensely. Moderation needs to be top, and for that, profits and engagement, which is why you see the Tinder-like model.

Online dating is now pay to win, how else would a dating site make money? It's not like the old days when throwing Google Adsense ads in between profiles could pay the bills.

My hetero friends tell me a lot of people they meet on apps are more interested in gaining OnlyFans subscribers than meeting people.

I don't know how prevalent this phenomenon is, but that's at least 0.9 on the Mad Max scale.

(As a demi/gay man, online dating is incredibly alien to me.)

It's true, at least 1/5 women on these sites is just trying to push a social media profile.

Some of them don't even bother with a bio, it's just a few photos and a link.

They are really easy to spot though. The ones that have no intention to meet and just signed up because of boredom, that's harder and wastes time.

Not only OnlyFans profiles, but also Instagram. Some want to get to a few thousand followers and use it, I know even some people in relationships that did it.

And then there's the bots which just start a generic conversation and recommend some Clash of Clans clone or something.

But as mentioned, you realize that in the first 5 mins of interaction.

Found my wife on OKCupid back when it was still good in 2011. Haven’t used a dating app since, but their decline makes me pretty sad, I have fond memories of using it.

Reply on news.ycombinator.com