Basically all the credit cards in Canada are now “chipped”, which is to say that there are visible microelectronics toward one end. To pay, you slip that chip into the reader, confirm the deal and enter your PIN. This allows for a surprising amount of variation in user-experience quality.

It’s a huge time-saver in restaurants because the little reader doohickey is wireless and they bring it to your table with the bill, thus avoiding one round-trip for the service person. [At this point, Europeans are wondering why I’m treating this as news, they’ve had it for years.]

Anyhow, there is an amazing amount of variation in the amount of work you, the customer, have to do accomplish the business of paying. To a point, this is reasonable. If there’s a tip in the transaction that’s at least one more step; and then some machines give you an option to tip by percentage.

The song-and-dance differs at nearly every business, and you’d think it’d be hard to screw up something this simple, but some of the user-experience designers out there have managed to make the process confusing, error-prone, and long.

The systems also vary in performance. Some dial up at what is, if I’m reading diagnostics correctly, 1200 baud, then wait for an overloaded mainframe. It’d be faster to count out the payment in pennies.

There’s a local liquor store that I don’t like much but is on the same block as our main grocery store, so it gets our business when I want wine or beer with dinner and lack the time to visit the nice bottle shop. They’ve made payment optimal: the cashier scans your bottle, hands you the reader, you stick your card in, it displays the amount and says “Enter PIN”: five strokes and you’re done. Also it seems to have fiber-to-the-reader broadband, it’s often accepted your payment while your PIN finger’s still in the air.

Given how much variation can infest a transaction this simple, and how badly it can be done, should make you think with more respect of the user-experience designers who try to make managing your privacy preferences or seat selections tractable.

I worship good UX designers myself; and the world has so, so few of them.


Comment feed for ongoing:Comments feed

From: ben (Sep 09 2011, at 22:53)

One is inclined to believe that it's not a question of "good" or "bad" UXDs, but rather a question of the extent to which they're being employed at all.

I've got a decade's worth of clients-not-giving-a-s--t to back this one up.


On the other hand, how many thousands of people were killed in fiery crashes before the air transport industry realized that they needed to pay close attention to human factors?

...So you can't say that this mistake hasn't been made before.


From: Michael (Sep 10 2011, at 01:58)

And also, people-who-know leave the tip in cash. If you include it in the credit/debit swipe, the processing corp takes their percentage of that, too.


From: Bart Schuller (Sep 10 2011, at 05:38)

Here in The Netherlands, we've had PIN transactions for a number of years, but they were combined with a swipe of the card to read the magnetic strip.

We've also had a separate system of "chip" transactions where you can put some money on the card beforehand and then pay small sums just by sticking the card into a reader and pressing "OK".

Both payment systems share the same terminals, which have a groove and a slot.

Now we're switching the vastly more popular PIN system over to also use the chip, like you describe.

When our local supermarket made the switch, it immediately became a usability disaster. To start, the groove for swiping had to be physically disabled (pieces of carton!) and instructions posted to please use the other slot. Then, the display on the terminal had to ask which system you wanted to use (press 1 or 2 and OK). The option for PIN used the unfamiliar term "Maestro" to maximize confusion.

To their credit, they seemed to have realized the mistake. After a week they went back to a few weeks of swiping and now again back to sticking in the chip. The confusing question is gone, as is the ability to pay using the "chip" system…


From: Richard (Sep 10 2011, at 06:22)

I see the tweaks to this process we've had for a year now haven't washed up on your shores:

* no need to insert the card into anything for it to register - just wave it at the proximity reader for purchases under a certain amount (currently $35).

* no need to enter a PIN up to the same threshold.

Together, these make quick buys from the shops a joy (or pretty close). It also makes things riskier, since a stolen card is now immediately usable in a lot more places (many fast food restaurants, all supermarkets, and a growing list of other retailers), as long as the thief doesn't spend too much at the same shop.


From: Peter Janes (Sep 10 2011, at 08:05)

Are there restaurants in Vancouver that actually have sufficient wi-fi to get to the tables? Here in southwestern Ontario, about half the time the servers will have me come to the cash register because the signal isn't strong enough to use the reader, and about half of the rest they'll have to do the signal dance, holding the reader above their heads, trying to find a sweet spot.


From: Harald (Sep 10 2011, at 11:51)

I've been ranting about the awful user experience with these things since they first came out. Some units display the transaction amount; others don't. Some make the customer tear off the receipt; others make the server do it (and some print them in the opposite order). And then there's the stores that have chip readers, but disable them because some other part of their workflow can't handle them, thus defeating the security advantages.

But my favourite blooper is the units that instruct you to press "OK" when they only have an "Enter" key (or vice versa). I even had one unit refer to the OK key in one message, and the Enter key in another!

This is basic stuff, and shouldn't require a fancy UX designer to get right. I despair for our profession whenever I use these things...


From: David Pitkin (Sep 10 2011, at 16:40)

Hi Tim, don't bash the 1200 baud modem it actually makes the transaction faster. The transaction data is so small and quick that the modem to modem negotiation actually takes longer with the faster computer modems. Now if there was a persistent connection that might be a problem but in this case the 1200 baud is a speed enhancement.


From: Steve Crane (Sep 11 2011, at 10:51)

We're still fairly new to "chip-and-pin" here in South Africa and frankly it's a nightmarish pain in the backside. There are the two systems; the older, swipe and sign slip or the newer insert card and enter pin. The problem is that businesses, or more likely the banks that provide the systems are having some trouble switching over.

If you have the older unchipped card, then they print the slip and you sign it but if you have the newer chipped card it gets more interesting. In a very small minority of cases you are asked for your PIN and nothing more but in the majority of cases you must enter your pin and also sign the slip. At the other end of the spectrum are a small number of cases where the slip is printed with "No signature needed" but they still insist you sign it. I suppose that last one is just down to poor staff training.

The weird thing is you never know which you're going to get. Just because two businesses have payment terminals from the same bank is no indication that the behaviour will be the same. I'm convinced though haven't confirmed this that it even varies from visit to visit at the same business.


From: Scott McIntyre (Sep 12 2011, at 10:26)

Do these systems handle splitting the bill?


From: Pies (Sep 14 2011, at 03:48)

"At this point, Europeans are wondering why I’m treating this as news, they’ve had it for years."

I was about to say :) The new thing around here, in Poland, is Visa PayWave, very similiar to NFC payments. I've seen quite a few shops accepting it, though, despite their advertising, very few in places I'd actually like to use it, like taxis or vending machines.


From: Unknown caller (Sep 21 2011, at 15:43)

Your current employer has recently sealed a deal with one such CHIP & PIN device OEM/ODM, perhaps you should direct some of your valuable time in lobbying internally for a good UX


author · Dad · software · colophon · rights
picture of the day
September 09, 2011
· Technology (87 fragments)
· · Software (71 more)

By .

The opinions expressed here
are my own, and no other party
necessarily agrees with them.

A full disclosure of my
professional interests is
on the author page.