This was the primary weak point in the implementation. Expecting users to recognize that they can click on the static-looking fields on the card (or relying on them to use tab to move between fields) does not pass the grandma test. It'd be great if you detected when a field had been fully filled out and moved to the next one automatically.

I think blind people / screen readers might have difficulty with this interface, too. It's important to keep accessibility in mind when designing futuristic interfaces.

If you press tab, it at least cycles through all of the input boxes on the front side in an appropriate manner, and you could probably design it to flip over on the next tab press. I'm fairly certain that would make it work with screen readers (but not positive).

It doesn't tab between the input boxes or even show the boxes until it's detected the card type based on first four digits. On initial load you get an input box with "XXXX XXXX XXXX XXXX" and nothing else. It does not look like an average credit card form and a lot of people will be confused "where do I put in my name" when they first see it.

I don't know. The current convention is always to enter the credit card number first, then the name later (usually as part of the billing address). Once you start entering your card, the rest of the information is explained nicely.

Yeah I really could have done with the tab order being a bit better. And flipping the card when tabbing would definitely help.

Those of us familiar with data entry get really used to tabbing.

It doesn't tell you if you go wrong either.

I surprisingly liked the idea and execution though (putting the other issues aside.)

Personally, sites that auto-tab to the next input drive me crazy. It makes it nearly impossible to edit if you make a mistake. Try explaining to grandma why the cursor doesn't stay where she put it!

I'm sad that I (a professional web developer for the past 7 years) failed the grandma test.

That's more of an indictment of the software than it is of you. FWIW it took me a moment too.

I'm forever failing to grok some websites, even though I've been working with and looking at them for years.

Showing real inputs until the user fills them in may be a nice compromise.

My experience on an iPad. Typed first 4 numbers of my VISA. Realized I could type in the MM and YY, then realized I could enter the name. It was a very smooth process for me - and I didn't read any text above or below before trying.

I wonder if trying it on a touch device made it more natural or not.