Sign Up Forms Must Die

http://www.alistapart.com/articles/signupforms

by jyu • 18 years ago

113 34 18 years ago

alistapart.com

jakewolf• 18 years ago

The irony is you have to sign up to leave a comment on that article.

khangtoh• 18 years ago

lol.. that's classic

Goronmon• 18 years ago

Except commenting isn't the primary purpose of the article. It would be ironic if you had to fill out a sign up form just to read it.

But I'm sure you were just poking fun, so poke away, haha.

serhei• 18 years ago

Use OpenID, perhaps? Add a ClickPass widget?

marrone• 18 years ago

OpenID currently only speeds up the registration process, not replace it

apgwoz• 18 years ago

But, how much "extra" information is really required for these services? Do they really need me to fill out my address or some other thing? Can't they just link my openid to a username of my choosing?

If that's the case, it speeds up the registration process by a lot, and _nearly_ replaces it. (not quite since I'd need a username)

petenixey• 18 years ago

You can completely replace the registration process as long as the application doesn't need any data from the user. If the app needs data then you can pre-fill it (as we do at Clickpass). Even pre-filled though you still need to confirm with the user that they're happy to send it (you can't allow an app to call any data it wants from a user without asking them first)

justindz2• 18 years ago

If the gradual engagement includes a reasonably complex operation, like writing an article and then creating an account to post it, does this obviate the need for a CAPTCHA?

I'm curious, because I hate CAPTCHAs, but only slightly less than I hate being used as a splog.

I think it might, so I'm feeling happy.

boredguy82• 18 years ago

Why would that remove the need for a CAPTCHA? Normal procedure: Register -> write spam. This method: write spam -> register.

justindz2• 18 years ago

Hrm. Fair enough. I was assuming the "write" step would include something atypical enough to screw up a program that just looks for "blogs" to "comment on." But, that might not be the case for everyone (or even for me).

mojuba2• 18 years ago

Bots won't be able to ab/use a site with gradual engagement, because the whole process would be too difficult for them to follow, I presume. Thus no need for captchas.

chengmi2• 18 years ago

That depends on the complexity of the bot. A general purpose bot might not get through a complex process, but someone could spend just a few minutes to customize a bot to a particular site. CAPTCHAs are reasonably good at stopping bots.

staunch• 18 years ago

How would you make this site have a gentler slope? Let the user post a comment and then say "Thanks for the comment, now you just need an account to publish this. Enter your username/password."?

Some of the sites that do "gradual engagement" piss me off when they let me believe I'm about to do something and then surprise me with a sign up form. It strikes me as tricky and I very frequently walk away.

I certainly believe it's smart to let users do as much as possible without requiring them to sign up. I think the idea of signing up is pretty ingrained in people though. If you make it dead simple (like this site or reddit) it's not much of a barrier and it keeps things straightforward.

gaika• 18 years ago

It is possible, and we have done it on jaanix. Take a look, there are no surprises.

You have a guest id with no questions asked that lets do everything on the site. You can keep it forever if you come back at least once every two weeks. You only have to join if you want to have your own subdomain - and we ask you for the name and a password - that's it.

andr• 18 years ago

Some users are used to registering no matter what. We only require registrations to upload content on the site, yet for some visitors the first thing they do on the site is register, without uploading anything.

TheTarquin• 18 years ago

"How would you make this site have a gentler slope? Let the user post a comment and then say "Thanks for the comment, now you just need an account to publish this. Enter your username/password."?"

An interesting idea, but it would probably deter casual commenters. People who drop by, see something interesting and want to add their two cents, sign their anonymous comment, and leave it at that.

It's a bit of a stretch, possibly, but track users and pester them gently if they comment several times anonymously. Something like "hey, I've noticed you've been around for awhile, if you sign up, you'll get these perks. Would you like to?"

http://qdb.us/latest does SOMETHING like this. It's a quotes database and after you've browsed around awhile it puts a non-intrusive nag-message at the top of the screen directing you to the FAQ. Maybe something similar to direct people to a signup form?

TrevorJ• 18 years ago

Seems harder to do with something as granular as posting permissions, I agree.

bayareaguy• 18 years ago

http://jottit.com has a nearly frictionless scheme.

jraines• 18 years ago

Yeah, it really hooked me -- even though I haven't used it for much, every time I have a simple, quick task where I just need to get a little info online fast I think of jottit

huhtenberg• 18 years ago

I am looking at their front page and I have no idea what they expect me to type in a very large unlabeled text entry field.

And if I click on Create, it say "see, that was easy". It certainly was, but what was IT exactly ? :)

cstejerean2• 18 years ago

Requiring users to sign up before they know if the service is valuable is wrong. Prove to them what you did is worth their time first, then require them to sign up if they want to continue using it.

You get a pretty good idea if HN is worth it by reading the posts and comments from others. If you like what you see and would like to contribute then sign up. Nothing wrong with that. It would be great to allow anonymous submissions and comments but unfortunately this is very likely to lead to abuse.

boredguy82• 18 years ago

"Upon arriving at [Google Video], they found a link to share their video and what happened next? They got the form in Figure 13.1."

"It’s only when you want to publish or share your movie that Jumpcut asks for your name and email so you and others can access the movie you just made."

I -hate- the web experience this article describes. If you're going to make me sign up to use your service, tell me. Don't string me along and waste my time.

The point the author seems to be making: let people know what they're getting before you make them sign up. Well, I think Google Video does a pretty good job at that. And I can -view- any video I want to without signing up. Is that "gradual engagement"? If so, why is it being criticized? If not, how is it different?

In fact, the more I think about this, the more the practice counters the principle. The principle seems to be (and ought to be), "Let your potential customers know what they're getting right away." Google Video does that. And if I ever want to go 'deeper' (from watching videos to posting them, or collecting them), I know right away that I need to sign up.

On the other hand, to _view_ the video I've spent the last however long _making_ on JumpCut, I get SURPRISED by a registration form. Plus, why would I need to register to -see- a video. I can understand needing to register to post or save a video, but to see it? I'd argue, then, that Google Video actually does a -better- job delaying the registration component than any of these examples.

And they're up-front about it from the start.

cstejerean2• 18 years ago

Google Video is not a good counter example to Jumpcut. I think Google Video works perfectly. The point of Google Video is to view videos, which it allows you to do without an account. To share a video you need to have an account, same as in Jumpcut. (Typically you have to make users create an account to post content otherwise a never ending stream of spam is likely to follow, degrading the quality of the service for legitimate users)

For Jumpcut, the real value in letting the user upload and edit a video is to get a feel for the video editing functionality without having to sign up, (since the video editing is really their selling point). If the user likes what they've created they can sign up to post it. This is key. Let the user try out the idea of your site without signing up. You can then require them to signup if they like it and want to keep their work.

Recently there was another post where somebody had a startup that had some cool way to learn a foreign language http://news.ycombinator.com/item?id=144998

I wanted to check it out but had to sign up to really see how it all worked. So I left.

mojuba2• 18 years ago

How does gradual engagement work precisely? If this is based entirely on cookies, which seems to be the only possibility, then: what if I started the process, posted something - my video, a blog post, etc - then got my cookies deleted? Or moved to another computer?

notauser• 18 years ago

It would depend on your product. I think the key UI principle to remember is that if user work is not saved, make it obvious that this is the case. They way they can decide when they want to keep their data enough to save it.

If the data is impersonal enough you can add a "Whoops, lost you work? Click here to see recent orphans" link.

mwmanning• 18 years ago

My current project involves interacting with a chat bot. We could require users to sign up on a form and make sure we have a boatload of data from them upfront before opening up the service to them.

Instead what we do is tell them to add a bot to their buddy list and send it a greeting. In doing this, they've already engaged in the service, and we have created an entry in our users database based on their buddy name and chat network. They can then decide whether supply additional information.

There are lots of ways to lower the barrier to entry if you're willing to change the way you think about online services.

naxxtor• 18 years ago

I think there is a fine balance between being "strung along" and gradual engagement.

For instance, if I were to use an app to share photos (a la flickr), and I was invited to upload a bunch of photos then be presented with "Ok, now you've got to sign up RIGHT NOW or we'll delete everything you just did" - that's sorta missing the point. That's less gradual engagement and more impending doom.

The implementation of the chatbot above seems to be more along the right lines, though. The holy grail perhaps would be allowing the user to fully engage with the service without even realising it.

chengmi2• 18 years ago

The issue isn't necessarily collecting "boatloads of data"; rather, the problem is making sure I have a way to contact this person who is potentially interested in my product. For you, this is conveniently solved through their screen name and network. However, how would you go about contacting someone who stumbles upon your website (without explicitly asking them for contact info)?

Another issue is accountability. If someone is abusing my service, I want to be able to ban them from ruining things for everyone else. If you look at sign up forms in terms of security, the psychological acceptability of filling out a form and providing an e-mail address is actually quite high compared to more robust ways of verifying identity (driver's license, social security number, etc).

jgill3• 18 years ago

If you really want to get rid of sign up forms, having a large scale trusted and secure central repository for unique IDs/logins and 'commonly requested' information would eliminate the need for many sign up forms. The majority of web based applications, services, etc. ask you for the same information repeatedly without any actual need or use for the information as a value add in the functionality of the application. Why is that? Why hasn't this been done/worked in the past?

aykall2• 18 years ago

Sounds like you are describing OpenID. Once i thought about a little software that would work like a real world id. Every time a web service ask you for personal info you would just click on a icon and show them your id (that means that the software would pass all the details to the service). Would that work? It sounds a lot like google checkout...

jgill3• 18 years ago

Yes, something similar to OpenID and Googlecheckout. The problem would be that the Yahoo!'s, Microsoft's, and AOL's of the world would not allow Google to centrally manage all user IDs and registration information on the web (or they would not be willing participants in the centralization of the information to another for profit company).

Sign Up Forms Must Die

Related Stories