I used to work at OkCupid. I could talk to you at length about the ups and downs and ins and outs of online dating, but the moral of the story I gleaned from my

I used to work at OkCupid. I could talk to you at length about the ups and downs and ins and outs of online dating, but the moral of the story I gleaned from my time there can be summed up pretty easily: People just want a ton of attractive/dateable people within a keyboard's reach, and if your site does any filtering on their behalf, people need to trust that it actually works. Most of the variations in online dating sites focus on optimizing pieces of that. The big guys (Match.com, eHarmony) make sure they get huge user numbers by campaigning on national TV, and they apply some simple internal filters that are 'approved by a doctor.' Smaller sites can't compete with that, so concentrate more on making sure you only run into people you'd actually be attracted to (see JDate). OkCupid's on the smaller side, and focuses on community building (for numbers) and uses mathematically semi-sound calculation coupled with quiz questions to create a user-driven but rigorous-feeling people filter. Basically everyone that's successful fits somewhere in that model. If I were you, I would focus less on specific features or trying to build a better datetrap. Instead, figure out what it'll take for you to get fifty thousand people signed up, approximately all at the same time, and then how you'll keep the site growing over time. If you get big numbers, the dating stuff will follow (MySpace...).