People keep telling me that dating today is a war zone, facing all kinds of challenges.

Dating apps don’t seem to be directly trying to help solve the problem as much as generate revenue. In fact, they are very directly motivated to not make great long term matches.

Some people seem think that just getting out there and hoping for the best is the answer. Maybe that’s true, but it’s still very random. I was wondering about a hypothetical alternative:

What if you could go to an agency of some kind get rated through a thorough evaluation process? Would that be helpful ? It’s not perfect, and many things are hard to measure. But maybe it’s a less random starting point and can escape the exclusively money driven approach of dating apps.

  • bouh@lemmy.world
    link
    fedilink
    English
    arrow-up
    5
    ·
    4 months ago

    The problem is not the how. The problem is that your agency is a company driven by profit. How do you make an agency that’s not driven by profit in a capitalist and consumerist society?

    • Skydancer@pawb.social
      link
      fedilink
      English
      arrow-up
      1
      ·
      4 months ago

      Easy. You either set it up as a nonprofit (still not great in terms of incentives) or, better, as a consumers (or members) cooperative.

      Alternatively, you don’t - you modify the incentives. Agency sets a one time, lifetime membership fee. Every failed match they set you up with refunds you a percentage of the balance.

      Detailed example:

      Let’s assume it’s a $1000 membership, 5% refund.

      First match works out? They keep the $1000 First match fails? You get $50 back. Second fails? You get another $47.50 (2% of the remaining $950) By match #45, you’ve been refunded 90% and they’re still holding less than $100.

      Why it works:

      This strongly incentivizes the agency to make the best possible match as quickly as possible. Users aren’t incentivized to join fraudulently because they’ll never get more out than they put in. The agency has no reason to create fake profiles, since a bad match costs them money.

      Where it (arguably) fails:
      1. This incentive structure is designed for long-term, monamorous relationships. It fails to account for poly relationships. People using it for short term hook-ups would settle over time into #2 below.

      2. After a certain number of bad matches, it’s not worth it to the agency to put any effort into making a good match. Since they make the most money on early matches, their incentive is to connect the most “desirable” candidates with new members. People with more failed matches will most likely be connected to … other people with more failed matches.

      Arguably, this is a feature not a bug. For new members, it means they don’t get spammed by long-time members that are hard to get along with or not actually looking for a long term relationship. For the ones that the early match algorithms didn’t work put for, it means they’ll at least get exposed to different groups of people over time - including others that failed to match for similar reasons as themselves.

      1. Even if the user never makes a successful match and gives up, the agency still gets paid. An alternate strategy for the agency could be to make the worst possible matches so members give up early and they keep more of the membership fee.

      This would not do wonders for their reputation and is probably not a good long term strategy for them - at least on the early matches. After a certain number of failures though, it might be an effective way to cut losses.

      • bouh@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        4 months ago

        Indeed. I’d add that you need the application open source and respectful of people’s data and privacy.

        Then there is another problem : you need people to adopt your application and use it. Because regardless of the qualities, the most important thing is the number of people using the service. That’s why Facebook is still not dead for example.