4 ms·
> but just because someone is rejected doesn't mean we didn't do our job well To my mind, yes it does. When someone is paying per match, if match that won't me
by shkkmo 12y ago
> but just because someone is rejected doesn't mean we didn't do our job well
To my mind, yes it does. When someone is paying per match, if match that won't meet you, it was a failure. I don't mind rejection, but I don't want pay to be rejected.
It doesn't have to go on indefinitely, but it makes sense to me to retry a couple of times that month and then offer the person the choice of either a refund or a new match next month with some advice on how to improve their profile.
- laurenkay 12y agoIf we did that, we'd have to charge $100/match. That's why we have Premium - it's a higher price point for a guaranteed date. But in order to personally match people and provide great customer service, you can't guarantee a date for $20. At least not yet, unfortunately.
- shkkmo 12y agoThat $100 pricing doesn't make sense to me. $20x2 per match is your baseline income per match here which I assume covers your costs. 1st try gets rejected, $20 bucks from the rejecter 2nd try gets rejected, $20 bucks from the rejecter 2rd try gets rejected, $20 bucks from the rejecter So $60 a month should cover it, assuming that you have a 100% rejection rate. $35 (.5 x 20 + .25 x 40 + .25 x 60) a month should cover it if you have a 50% rejection rate. So assuming you have a 50% or less rejection rate, I would pay $35 a month for at-least a 7/8 chance of meeting someone every month.
- shkkmo 12y agoUsing a similar formula, $100/match to provide a guaranteed date would mean that your rejection rate is about 83%. Estimated using this forumula: (5 = x^1 + x^2 + ... x^n) and solving for x as the probability of rejection. I used an n of 29 for the estimate.
- csdrane 12y agoHow'd you derive that formula?
- shkkmo 12y agoPoorly, I apparently missed the first term of X^0 $100 = expected cost of matches $20 = cost on this side per match X = probability of rejection (which leads expected cost = cost of first match times probablity it is needed plus cost of second match times probability it is needed... plus cost of nth match times probability it is needed. 100 = 20x^0 + 20x + 20x^2...20x^n 5 = x^0 + x^1 ... x^n The actual number is more like 72% :/