Comment by dexen | Hacker News Reader

Comment by dexen | Hacker News Reader

Parent Full thread

>other options are substantially worse

Talk about misaligned incentives!

The dearth of effective dating platforms is the expected end state of the current "pay per use" model. The popular sites earn more from repeat customers than from single-time customers. That is fine for general goods & services. However it goes against the usual idea for dating: to find a partner & establish a good, lasting relationship in as few "attempts" as possible. Users successfully coupling up are effectively "lost" for the platform that facilitated it.

What would it take to build a sustainable dating platform? First up, it would need to benefit the most from a successful & permanent match, rather from repeated use.

As much as I dislike Facebook for their abysmal handling of privacy, I figure FB Dating suffers less form misaligned incentives than standalone dating platforms. Presumably users will remain on FB proper anyway, whether their dates are successful or not.

Come to think of it, the "traditional" (for lack of better word) society built on nuclear families just might have been geared that way - to benefit the most from people finding partners & becoming long term couples.

>What would it take to build a sustainable dating platform?

Since finding the right person is a numbers game, I'd say the best path is an easy path. Something between speed dating and a normal date, where the goal is to check for chemistry as well as advance past the initial, in-person interaction. Start with having a few like/dislike parameters met, then get you both face to face.

Here's my idea for that:

https://news.ycombinator.com/item?id=17356401

If we simplified and just went for a real-world, slower version of speed dating, where each pair would get 15 minutes with each other, it wouldn't control for like/dislike parameters. IE: I definitely wouldn't date a smoker. Having it start online then move to real-world, asap, helps to match likes and dislikes more closely.

If you don't mind bit of a downer, I see three major obstacles in your idea:

1. monetization. How do you ensure your platform can at least cover its costs? The users want to know your platform wont go dark one fine day because of an overdue server bill. Better yet, that it won't get yanked like many of Google's cool projects due to insufficient profitability.

2. incentives alignment. How do you ensure the incentives for your platform are to get the users paired up permanently, which usually would mean they stop perusing your platform? Unlike typical services, a dating platform is one such that most people want to be able to ultimately do away with.

3. fulfillment. How do you make users feel they are putting in reasonably hard effort? How do you let them increase the effort if they so desire? People do feel fulfillment from putting in effort; cf. a lot of computer games that grant satisfaction through daily grind.

[edit]

A throw-away thought:

current dating platforms operate in "push" model: the singles make themselves available.

How about you make a "pull" platform - let already established couples find & match up great candidates to join their social circles - as new couples.

Making the already established couples your main users would solve the obstacles 1, 2, and 3. You would take money from the already established couples for finding & matching up new great, interesting, friendly couples for them. You would have the already established couples "grind" (put in the effort) for finding & setting up new ones.

The idea doesn't sound like it would require much monetization. It isn't a site you actively browse, it just makes matches and dolls out notifications. You could probably run the entire operation for a major metropolitan area on a raspberry pi.

> Talk about misaligned incentives! > [...] > Users successfully coupling up are effectively "lost" for the platform that facilitated it.

It works the same in healthcare. If we are cured once for good, they charge us once. If there are good prophylactics and we stay healthy - even worse.

Yet, somehow the system works, to a certain degree. (Why? I don't know. Maybe there is more stigma with destroying someone's health for monetary gain.)

...

Maybe there should be a service, for which we pay ONLY during times when we are healthy (or have happy romantic/sexual life)? In that way, the incentives would be aligned.

Any ideas of how to do so?

...

As one remark: a lifelong relationship is not the only goal of dating. Some people look for a short-term, or even an ONS. Also, if a relationship goes bad it is better when it does not stays for a lifetime (in the worst-case scenario, when one's life ends with a murder, or - suicide).

> Yet, somehow the system works, to a certain degree. (Why? I don't know.)

Well for starters, there is still some level of competition in the pharma industry. If you're trying to make a better version of a competitor's product, changing "It makes X go away for 2 weeks" to "It cures X" is a good strategy.

Secondly, it's not as though we can program treatments to act in exact accordance with executive whims. We use what works.

(To be clear, none of this is to say there isn't a problem. I am listing reasons the situations is not as disastrous as it theoretically could be.)

> The popular sites earn more from repeat customers than from single-time customers.

> What would it take to build a sustainable dating platform?

Between "earning more" and "sustainable" there's is a broad spectrum of possibilities. I know from past experience that it's absolutely possible to skip many deliberate ways of "earning more" and still be sustainable (even highly profitable). Perhaps you won't conquer the market without shady tactics nowadays when competition is fierce, but you certainly don't have to screw your customers to be highly successful.

It shouldn't be that hard to set up a dating site with aligned incentives - I had the idea of a site which requires a lot of personal data on signup, and deletes it for a fee at account deactivation . People pay huge premiums for matchmaking agencies, building an algorithm with similar incentives shouldn't be that infeasible...

Reply on news.ycombinator.com