Signups increased by 60% after actually removing the signup form | Hacker News Reader

Signups increased by 60% after actually removing the signup form

visualwebsiteoptimizer.com

80 pointsparaschopra22 comments

It's not very surprising. The original page was not very good.

The form was too prominent. It shadowed the product description.

Furthermore, putting it on the left made the conversation/narrative backwards. You first convince people to sign up, then present the form, not the other way around.

The second design corrects both of these mistakes, but correcting them without removing the form would probably have improved the conversion rate as well.

I agree. This reminds me of the book Why we Buy, about the buying habits of retail shoppers. When shoppers walk into a store, there needs to be a threshold where they can acquaint themselves with the surroundings. Anything that is too close to the entrance gets ignored by most shoppers as they walk in. Sites with sign-ups on the front page feel that way to me. I haven't had a chance to see what the product or service even is yet, and they're already pushing me to register. I like a little front matter before making a decision.

Interesting analogy. And this is what veteran internet marketers keep doing with the never ending sales page (for parrot training kind of ebooks). Have you noticed how they have call to action at the end of the page and never at the top?

Actually, they repeat the call to action many times throughout the page, in the long sales letter format.

Yes, but it is never at the top (before making a case for call to action).

Yes, agreed. The form was bit too overwhelming on the original design. Plus it said 'Vendio Store' instead of 'Ecommerce Store'.

But I am curious how would you have preserved the original form in variation? If you had put the form, it would have again competed for attention with product description. What variation did here is classic example of "squeeze page".

The more I think of it, the more I think that, indeed, the page is better without the form.

An interesting variation would be to expand the "Signup Now" button into a registration form with Javascript. It would greatly reduce latency and perhaps give a better impression to the users.

What happened to the page described in the article? I couldn't find it at http://www.vendio.com.

This is the page that users got to see once they clicked Signup now: https://secure.vendio.com/ecommerce/litelogin/hosted_store

Agree. The form itself is awful, presenting the user with 5 fields to fill in, and two checkboxes before they've made a conscious decision to sign up. Having a mailing list button on by default and setting it apart from the main page text also makes users inclined to think it's for a mailing list rather than the product itself, with the vague "Vendio Store" heading not exactly helping.

Spamming sites looking to harvest emails look a lot like the first in terms of presenting a form. That's the first thing I notice on the first page. The act of clicking on the button means the user as committed some effort towards signing up and is more likely to continue whereas on the main page even if they have chosen to start the form they haven't quiet committed in the same way.

This reminds me of "Sign Up Forms Must Die" article by Luke Wroblewski (http://www.alistapart.com/articles/signupforms/ with some additional resources here: http://www.lukew.com/ff/entry.asp?1219)

Maybe it was because the "100% FREE Ecommerce Website" title was much larger and prominent in the non-form version.

Yep, also not quite the same verbage. I would put most of the weight on the change to this headline, as well as its better styling with the whole website

After reading some number of such stories I come to conclusion that you should tweak your user interface until you discover drop in signups (or other important parameters).

Basically it amounts to manually perform hill descent algorithm. ;)

But that will only be optimizing for local optimum. What you need is a shake-up from time to time and that means testing radically different ideas!

I think the reason for this is our innate hatred for signing up for things(work). Putting this right in front of you may make it easier and quicker to get to the function, but: "uhg another form"

Provides no competition to the back arrow. Made me stare at the form, and flick my eyes around vaguely for any mention of purpose or benefit. Very poorly done originally.

The original was very off putting, it almost looked like a magazine subscription flyer. I can see why the revised version performed so much better.

Could it be a statistical fluke? I'm very skeptical about reading too much into A/B results without variance/confidence interval estimates.

It wasn't a fluke -- variation was statistically significant better than control (we never publish results otherwise). Clients are never happy to see their actual conversion rates and variance, etc. published in the real world, that's why I choose not to write about those figures.

I can't say this with any authority, but looking at it, having the sign up form in the beginning is not good. They might have better results with an embedded signup if the signup form is at the end of the page.

Also the background of the signup form is a terrible dark blue color, what's the color of the signup form on its own page?

It's white. Here is the form which appears after you click on Signup page: https://secure.vendio.com/ecommerce/litelogin/hosted_store

Reply on news.ycombinator.com