Thought experiment: How many people never sign up at all because you ask for email and password twice each?
Did you track how many fewer signups you get now that you have that double email? Are you net positive on the total activated users per day, adjusted for growth?
I have never aborted signing up because the form asked for the same information twice. Nor can I imagine doing so if it was obviously intended.
I could've done A/B testing and I could've done a full statistical analysis of the results. But I don't think splitting hairs like that is particularly useful. I simply had no reason to believe that it would affect our growth negatively, nor do I have such reason now. We are growing very, very fast.
But what I was really referring to was the customer service overhead. Almost every account that had a failed e-mail address would cause us some customer service. People creating another account which needed approval to prevent multiple accounts per user (which is cheating the game), or just asking questions. I cannot imagine people asking less questions if they paid for an account and can't access it.
Did you track how many fewer signups you get now that you have that double email? Are you net positive on the total activated users per day, adjusted for growth?