5 ms·
Why the downvotes? A better dating site is on YC's list of things they want to fund.
by gaius 17y ago
Why the downvotes? A better dating site is on YC's list of things they want to fund.
- jrockway 17y agoWhat's wrong with okCupid, other than being written in C++?
- randomtask 17y agoWhat's wrong with it being written in c++?
- jarin 17y agoIt's not written on punchcards, that's what!
- philwelch 17y agoIt doesn't have the best revenue model and it hasn't fixed the fundamental gender imbalance problem yet. These two problems have the same solution: charge men for membership, women join for free. Every college keg party has figured this out, I'm surprised online dating sites haven't.
- cabalamat 17y agoDating sites run on the network effect: they are useful in proportion to the number of people on them. If they charge money for members, they will have less people on them. Therefore someone interested in a dating website should join one (or more) that don't cost anything. Oh, and they are less likely to be scams that way too.
- jseliger 17y agoBut the quality of dating sites matters too: is a site with 1,000 flakes or 200 people seriously looking for a date better? One thing that charging, even very small amounts, can do is weed out people who don't care. For example, Love Lab: http://thestranger.selectalternatives.com/gyrobase/Personals/ http://thestranger.selectalternatives.com/gyrobase/Personals... through Seattle's The Stranger seems to work pretty well because a) it's already oriented in the community and b) charges, which keeps out spammers and flakes. You might be right, but I 'm not sure which element is more important: sheer numbers or the quality of the people who have joined.
- gaius 17y agoExactly. On a site that men pay but women join for free you're going to find a lot of women who expect men to pay for everything for them, and the only men on that site are going to be the kind who expect to get what they pay for. Now, I'm not saying there's anything wrong with that, everyone's a grownup after all. But that isn't the demographic that match.com et al are after - they measure their success in how many marriages they produce. Marriage-oriented people are also more likely to be at the point in their lives where they have money to spend to, they're a highly desirable market.
- cabalamat 17y agoHow do you define quality? What you look for in a date might be different from what I look for or what someone else looks for. In fact, it probably is. The more people on the site who live near to a user, the more likely that user is to find someone compatible with them. Having said that, "quality" might be important if it's a way of making it likely that users are more compatible with other users, e.g. religious-minded people who want to marry within their religion.
- philwelch 17y agoFor the heterosexual use case (90-95% of use cases?) the network effect is a two-variable problem. You have to consider not only the number of eligible mates but also the sex ratio, since the sex ratio affects your odds-per-eligible-mate. If there's more men than women on the site, after a certain point the network effect for men actually gets worse as more men join. It's not that great for women, either, since they're overwhelmed with contact by large numbers of men.
- potatolicious 17y agoI dunno - bars/clubs haven't really solved this either. Sure yeah, you have things like no cover for women, but where I am the ratio is still not too awesome. This hasn't really stopped men from trying, though. The gender ratio doesn't have to be perfectly balanced, all is has to be is not ludicrously out of whack.
- philwelch 17y agoClubs can't programmatically adjust rates based on the current balance, though. The best they can do is have bouncers to filter the quantity and quality of men they let in. A bouncer can be pretty clever but there's room for an algorithm to do well too ;)
- maukdaddy 17y agoIf you're worried about what language a dating site is coded in, I'd suggest not signing up for any dating sites =)
- luminary 17y agoI actually wrote about this the other day: http://whislr.com/2010/04/20/a-different-approach-to-building-dating-sites/ http://whislr.com/2010/04/20/a-different-approach-to-buildin...
- mseebach 17y agoIf partner selection was as simple as being hooked up and talking to a total stranger for a few weeks, there would be no dating-sites, and blind dates wouldn't be dreadful. There's a reason first dates are often for a early-evening drink in a bar - if just talking long enough would do the trick, first dates would be in a secluded cabin in the mountains :). I like your thoughts on targeting young professionals, and basing it on an existing network such as LinkedIn also sounds reasonable. Your focus on keeping personal information out of the conversation is not viable - remember, it's not just e-mail and phone numbers, it's also suggestions to meet. I also think that the fee to be hooked up with a second person is going to be a major deterrent.