How I increased conversion on my checkout form | Hacker News Reader

How I increased conversion on my checkout form

pud.com

71 pointswasd21 comments

Noting that on the before the "you're one step away" immediately jumps out at you.

Personally I don't like that it seems like overselling to me. I'd like an a/b test just with that one phrase actually.

The ideas presented sound good in general and make sense. I'd guess you'd have to tease out which of them actually did the trick.

Anyway, another possibility is that there was something negative on the sparse form (like the "you're one step away") and by adding a few things to the page it detracted from that.

In the same way that if you have a big nose it's less obvious if you do your hair (and makeup if a woman) in a certain way. So it's not the fact that you are adding makeup or hair as much as it is that the nose doesn't appear as big.

Now that these items are all in, it might be interesting to see him run a multivariate test to see which of the factors actually contributed to the 60%, and if all of them are net positive.

I'd like to see performance vs. Stripe Checkout, which we use. Like this: https://stripe.com/img/blog/posts/checkout-new-features/blog...

I would also like to see any analysis of which of these many changes made the difference. Did only particular thing account for most of the difference?

I'd like to see performance vs. PayPal.

I launched a project a month ago (a desktop Windows app) and only had the Stripe checkout. Guess what? No one outside SV and HN has any idea what the heck Stripe is and half of them "don't feel like giving you my credit card info" even though it says right there that it goes directly to Stripe. So after 2 weeks I grudgingly added PayPal and added itsy-bitsy hard to find link to it on the checkout page... Drumroll... nearly 50% increase in sales!

This post makes me sad, but reminds me just how big of a technical gap there is between the valley, the early adopters across the country, the rest of the population in the middle, and then the older/lagging adopter generations at the tail. (I'm picturing a normal bell distribution here).

A lot of companies are creating solutions to problems the majority of people don't have yet, and I think the gap is widening.

Bear in mind it wasn't long ago that nobody had any idea who Paypal was...

Adding a credit card type dropdown seems unnecessary. There are standard number patterns for each card provider, and making users select one is silly since cc gateways don't typically ask (they use the number patterns). I'd like to know if something about adding that step actually helped conversion or had no affect at all.

I think it helps the user feel confident in the form's accuracy.

In some cases, abstracting or automating away information from the user can make them confused or increase distrust in the form. There's something to be said for making things 100% clear and upfront to the user.

This is something you'd test, it's not something I'd say with confidence would increase or decrease conversion rates; but I definitely understand the hypothesis behind it.

Alright, I'm fucking done with HN comments. So damn tired of being down-voted over nothing. It's like a hive of 5 year olds.

I'm going to guess that you got downvoted as people felt your comment didn't add anything to the discussion, which is the prescribed use for downvotes. The article already explained the rational for adding it and that it wasn't actually used. It's likely clear to you why your response was also downvoted.

I got downvoted -12 a few weeks ago for saying that an article mentioning John Regehr was the professor from the University of Utah that I respected. What the fuck is that? This website is overrun with nerds with small egos.

You said "I saw Regehr and knew it had to be the U prof."

How did you expect people to respond to that comment? The comment adds nothing to understanding of the article. The comment adds nothing to understanding of the author of the article.

That's true for half the comments on any given article. I call bullshit.

Great info. It's beyond the call of duty but wonder if Pud or anyone else could quantify any of the individual changes?

Why does the before form say "Secured with 256-bit SSL encryption" when the after form says "128-bit SSL"?

He mentions that he was inspired by this article:

http://baymard.com/blog/visually-reinforce-sensitive-fields

From which it looks like he copied the "128-bit SSL" text verbatim.

It's a bummer that all of these were shipped at once. Tough to tell what the true takeaway is from this (security vs. reinforcing value proposition). I'm sure this was quickly put together to ship, but it would have been nice to have distinguished the valuable changes from the red herrings.

> Added card-type dropdown ... Their selection isn’t stored or used for anything.

I wouldn't have thought of on my own, adding extra fields that do nothing except increase conversion.

> adding extra fields that do nothing except increase conversion.

Well, that's not certain. Any of the other changes could be responsible for the increase, and not this particular change.

Compared to how it looked before - great job!

Did the strength of the ssl decryption change? Do some browsers not support 256 bit ssl encryption?

No kidding; I thought AES256 was present even in SSLv3.

Reply on news.ycombinator.com