User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It only takes a minute to sign up.
Sign up to join this community
Recently I've noticed more and more sign-in / sign-up workflows defaulting to the sign-up option first and providing a minimal "Sign-In / Login" action somewhere else on the page. What's the reason behind this practice?
Examples where I've seen this recently:
I agree with Igor-G that one conversion type (getting a new signup) is more valuable than the other (an existing user logging back in), but I think the primary reason is slightly more complex, and would still make sense even if the two we're of equal value to the company.
It's a function of how much each user-type's conversion likelihood can be influenced by the defaults to complete the action.
Essentially, you're deciding which user-type has an additional hurdle imposed by the default choice (they have to find the "log in" or "sign up" toggle or link that wasn't the default, and click it). So, even if both conversions are equal, you want to optimize the default for the user who is most likely to crash into that hurdle and quit, not the one who is more likely to succeed even with the hurdle.
Returning users are more motivated to clear hurdles, as they have come back again to a thing they deliberately signed up for. You know they are motivated to get into your site, unlike the new user, who is just browsing. (Or lost. Or bad at typing urls.)
Returning users are more qualified to clear hurdles, as they're presumably a bit mor familiar with your site in general, and have already demonstrated an ability to fill in a bunch of fields correctly, and probably to click a confirmation in their inbox. It sounds silly, but when you look at form attrition, it's fair to say that someone who's completed a sign-up can be assumed to be a good bit more adept at dealing with any extra step than a random visitor.
Returning users have fewer hurdles downstream of this one, so adding one results in a lower maximum number of hurdles. The returning user who's given the extra step of being forced to navigate to the "log in" page likely only has to deal with two fields to be done. The new user likely has at least 3 plus an email fetch quest ahead of them. (The Facebook form has 9 plus email). So, even though adding a step to login means a user has to do it many times, the total annoyance that might cut off any given conversion is much lower, and thereby presumably farther from a given users "oh, f*uck it" point in this path than the sign up one.
I believe this happens because of the few reasons:
I don't have numbers on that, but from personal observation and from what I have heard other designers mention, people don't log out. They register once and don't want to enter yet another username and password, so they check "remember me" on their first login. This is certainly not true of everybody, but I think this is how a vast majority of users do it. And the websites are certainly obliging enough to leave them logged in for very long time.
And when the "remember me" time is up after a few weeks and months, they don't go to the landing page. They go to the address suggested in their browser URL bar, which is typically a page customized to the user, such as "my profile", "my inbox", "my deals" or whatever you have. When the web application has a non-authenticated request for such a page, it usually automatically redirects to a login form.
Due to this user behavior, I think that almost everybody who arrives at the landing page without already being logged in does not yet have an account. If companies have measured it and confirmed the numbers, it sure makes sense for them to optimize the page for the majority of people who sees it, that is, the not-yet-registered ones. The fact that a registered user is more motivated to log in and doesn't have to be lured into it is certainly a nice circumstance, which makes it even more of a no-brainer.
