2 ms·

Hmmm... Only providing one password field seems like a road down which madness lies, even if it's unmasked as here. We had an app where we'd ask for e-mail add

by lucumo 17y ago

Hmmm... Only providing one password field seems like a road down which madness lies, even if it's unmasked as here. We had an app where we'd ask for e-mail addresses. A lot of people mistyped their e-mail address, which was of course unmasked. As much as half of the non-activated accounts were do to a typo in the e-mail address. We now ask for the e-mail address twice, it made activation failures a lot less likely.

jasonkester 17y ago
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?
- lucumo 17y ago
  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.
  - jasonkester 17y ago
    I only mention it because users actually do behave strangely this way. Every field you show them loses a percentage of them. It's weird. Here's an experience I had a couple years ago where a small tweak to remove a signup barrier more than doubled the number of people signing up: http://twiddla.blogspot.com/2007/04/1000-signups-on-day-one.html http://twiddla.blogspot.com/2007/04/1000-signups-on-day-one....

by lucumo 17y ago

jasonkester 17y ago

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?

lucumo 17y ago
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.
- jasonkester 17y ago
  I only mention it because users actually do behave strangely this way. Every field you show them loses a percentage of them. It's weird. Here's an experience I had a couple years ago where a small tweak to remove a signup barrier more than doubled the number of people signing up: http://twiddla.blogspot.com/2007/04/1000-signups-on-day-one.html http://twiddla.blogspot.com/2007/04/1000-signups-on-day-one....