The standard name for this design pattern is Lazy Registration. I admit Progressive Signup is probably a better label as it is more descriptive, but lazy registration is what people have been calling it for some years now...
Naturally the question I was left with at the end of the post was "Does this impact user conversion and retention in a positive way?" thus leading to "Does this impact revenue and thus profit in a positive way?"
For those interested in this. Luke Wroblewski wrote on a similar theme "Gradual Engagement Boosts Twitter Sign-Ups by 29%" - http://www.lukew.com/ff/entry.asp?1128
Yeah - this would annoy me, but I would class myself as a "power user". I just want to get in, give you my email, password, confirm password and be done. However, as @aeden mentions, what I think doesn't matter if it converts better!
Also known as a neat trick you can try out on hobby/free apps but are most likely horribly inappropriate for the average startup. As soon as you get even remotely serious, for example spending money on marketing, you're going to wish you captured email addresses up front. When you're spending money on customer acquisition, "cool, 10,000 people tried our app" doesn't mean nearly as much as "cool, we have 2000 email addresses".
> As soon as you get even remotely serious, for example spending money on marketing, you're going to wish you captured email addresses up front.
Just to add a dissenting view, I am ruthless when it comes to who gets (or, usually, doesn't get) my e-mail address.
If you require one at any stage before I am ready to commit to your company, you will almost certainly lose me as a customer forever.
If you take one and ever use it for anything other than the explicit purpose for which I gave it, including sending unrequested marketing spam, you will almost certainly lose me as a customer forever the first time I receive such a message. You will certainly get killfiled so you can't reach me again in future.
The only exceptions are likely to be established companies with which I have some sort of history, if they are sending something that might be genuinely interesting, e.g., when Amazon started sending out recommendations e-mails and gave a clear way to opt out right from the start. If you are a start-up, it is highly unlikely that you will be in this category, however.
[edit] We have an email alert feature on our job board platform, that allows job-seekers to get email notifications for new job ads. Standard functionality. When we removed the registration barrier from this user journey (so it's now just a single text field and submit button), email alert creation rates basically doubled on most client sites.
Anything that reduces friction for a user to get started with your product has the potential to be a good thing.
But if your goal is to make money, and not just generate usage, I can see some possible problems with this approach. Not having a way to circle back with a user is bad.
Also, if someone truly needs a solution to the problem you are solving a signup form will usually not stand in their way.
It would be interesting to A/B test the entire user lifecycle to see if this approach has impact on the metrics that actually matter.
To me, its not just about lazy registration. It's about the mindset of the user when you ask them to do something.
For example, when they have created some content and want to save it, that is the time to ask for email/password. It works better because of the implication: you are saving their work.
Suggested feature: Implement an RSS feed of each user's published writings. (Or post these articles on a blog.) I'd like to receive future updates without having to watch HN. I'm guessing there are at least a couple of others here who would like to do the same.
I prefer Costco to Best buy any day. What matters far more is reputation and the quality of the product than the barrier to entry to join/utilize the product.
Facebook could add a ten page signup form and I doubt it'd substantially impact their legitimate signup rate.
I suspect that Lazy Registration/Progressive Signup is far more important for lesser known sites. Facebook has the weight of some 500+ million users behind it and there is sufficient motivation for new users.
When working on wedding proofing sites I found that users would be willing to put up with more steps because they were motivated to begin with.
What about openID integration? give the users an option to sign up using their google-id or their facebook-id or any of the other id's the user already has. I would prefer to do that over giving my password to another site.
http://www.google.co.uk/search?q=lazy+registration+demo