Dating apps have become enshittified and is why they are so ineffective. However, there is a huge opportunity for improvement. The core hypothesis my app is: a pay-per-mutual-match model that aligns user and app incentives. The mutual financial investment also makes users more likely to engage once they match.
My app doesn’t rely on complex algorithms or AI for matchmaking, as they perpetuate biases. Instead, I use a transparent system where users fill out a detailed questionnaire, and I match them based on their answers using SQL queries.
The app is built with open source software (Docker, Django, Flutter, and more!), so no vendor lock-in. This is the first app I ever made so it has been an incredible learning experience.
Ask me anything about my app, Qtpi, my tech stack, or my journey!
Troll was a poor choice of word, but there is lots of antisocial behaviors that a user would be disappointed with having to pay for. I.e: dick pic, ghosting, cat fishing etc.
Refunding does sound like a reasonable answer though, so at least you’ve thought about it. Good luck!
Thank you! As of now, I don’t plan to support sending photos for that very reason.
I also plan to track ghosting eventually by having the ghosted user push a button that starts a stopwatch. Then it notifies the ghoster that they are being timed. They can either reply or unmatch with no response. In either case, closure is gained. There could be minor consequences like putting the average reply time publicly on each user’s profile too.
Of course, this is just a tentative plan for now. I have higher priorities to implement than this currently.
This is interesting. Hinge pushed out a message to users that was like “studies show that ghosting makes you and the person you’re ghosting feel bad! It’s not a big deal to just say no thanks!”. I don’t know if it made a difference.
One of the things to take into account is some people (often but not always men) are dangerous. Women have told me they’ve ghosted because the man was giving them danger vibes. Encouraging the person to unmatch might be good enough.
One thing I’d like to see, I think, is some way for me to rate people I match with but didn’t keep dating. Like Alex was polite and punctual and respectful, but they hate cats so we can’t be a thing. All that info just goes into the void. Though i wouldn’t have this be free text, and there’s more ways to screw it up than do it right. Information about people ghosting or flaking would go in the same bucket.
Yes, those are excellent points. I have thought about that and the implementation for the following is in the works:
If you’re familiar with the web of trust like in gpg or other networks, it’s basically transitive trust. If user A trusts user B and user B trusts user C then user A trusts user C.
I am going to have the ability for users to verify each other when they meet in person. It essentially says, “this person matches their profile”.
The next level of trust is “connect” which is analogous to casual friends and acquaintances. So if you meet someone but you don’t fit, you can connect and give some “social proof” that this person is alright. Or you can do it for your coworkers or whatever.
The final level is “trust”. It is for close friends, close family, etc. It gives stronger indication for this user.
All of these can then be recorded in a graph database. Then when a women (for example) matches with a man, they can see how many people verified, connected, trusted the user and how many “edges/degrees of separation” they have with this user. This means that a friend of a friend could put the user more at ease. Or if at least a handful of users verified the profile, there is less chance of a catfish, etc.
Yes, that graph sort of thing is exactly something I’ve imagined before. I think it has potential. Assuming you have users, that is.
Thank you! Yes, I think it has potential too. I do hope that I am able to get users. That will be the crux of the problem. I’m going to start trying to market and get college students to try it out. We will see. Hopefully it goes well.