I am working on a registration application which would be used in a live "museum" type experience. Essentially, to enter the museum you need to register for a personal account to receive a ticket.

I am considering tapping Facebook/Twitter/Google APIs to decrease the total registration time per user by pulling their information through the respective API. I can find information relating to using the APIs for website registration however my question is related to abandon rates when offering this on a public terminal.

Does anyone have non-anecdotal evidence which supports or denies increased abandonment rate on public (non-personal) devices when requiring Facebook, etc account information?

  • 6
    There is plenty of evidence that registration increases abandonment when done in private. Considering the disadvantages of a public terminal (inherently harder to use an unfamiliar system, public systems are less secure, more people are looking over your shoulder, etc.), why do you believe public users would be more willing to register in public than in private, where it is already known that registration represents a barrier?
    – Brian
    Jul 16, 2013 at 21:17
  • The ticket itself could be used as ID. Then users could use it to register at home after visiting a museum. So maybe you could change the flow, it is unclear from your post. Jul 17, 2013 at 6:48
  • is it a UX related question?? i am new to here can anybody please tell me what is the purpose of this site.
    – IamAbhijit
    Jul 18, 2013 at 6:40
  • @viewPagerSpecialist please see tour and help center pages and the User Experience Meta site for any questions you have about the site!
    – elemjay19
    Jul 18, 2013 at 20:18
  • 1
    @JamesJenkins: Just Google "shopping cart registration increases abandonment" (sans quotes). Support for this claim is ubiquitous. One of the most well known examples is probably The $300 million button.
    – Brian
    Dec 20, 2013 at 17:41

2 Answers 2


I am afraid I don't have a definite answer but the only way to find out can be by doing some testing and finding out if users are willing to use Twitter\Facebook\Google login credentials. The two user research studies I found show that many users may not be willing to use those login options if an existing login already exists.

To quote this article from mail chimp

I did a little extra analytics footwork to see just how many people were clicking the social login buttons. I was shocked to see that just 3.4% of the people that visited the login page actually used Facebook or Twitter to log in.

Similarly this study by typepad has similar results and shows that most users prefer to use the type pads login option as opposed to other open authentication options. To quote the article

To find out which services people are using to log in, I contacted Jonathan March, Six Apart's Data Architect. I found out that while we don't track every log in specifically, we can get a very rough estimate by looking for a URL in a user's ID. Though the data is a bit out of date, I think it is still very useful.

enter image description here

That said, this data is based upon the fact that users already have a login and people might not be willing to try a new login option even though its a social login. To continue with the study, the study states that surprisingly most of of the new logins came via open authentication. To quote the article

What about sign up?

The previous graphs are based on data about users who have LOGGED IN in the past few months. What about new users - users who have SIGNED UP recently? Luckily Jonathan thought of this and followed up with data and graphs on the growth of alternative authentication providers. The results are a bit shocking.

User sign ups between 2 and 3 months ago (June 28 to July 28)

enter image description here

User sign ups between 1 and 2 months ago (July 28 to Aug 28)

enter image description here

User sign ups in the past month (Aug 28 to Sept 28)

enter image description here

OpenID user sign ups by provider by week

enter image description here

That said, I recommend reading this article about which open authorization option to choose if you decide to go that route.


Like others, I do not have hard data for you. But I run several online services, and requiring registration is absolutely a barrier. If you are going to require registration, allow users to use an existing login (e.g., fb, twitter or g+), and don't ask for that login until the end of a demo period.

  • 1
    Welcome to UX.SE! Check out the help center and tour pages to learn more about getting the most out of this community! OP specifically asked for non-anecdotal evidence, which your answer does not provide so it's not really an answer to their question.
    – elemjay19
    Jul 23, 2013 at 15:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.