QR Code on a Philly Cheesesteak

Take a look at the image on the right. Someone in the Johnny Rockets marketing department decided that their customers would be so excited about the new Chicken Philly Cheese Steak that they would download a QR scanner app to their phone, launch the app, snap an image of the code and be amazed that the phone magically took them to… wait for it… www.johnnyrockets.com.

But there may still be hope for the awkward QR code. The key is to find high-value processes that are a big enough pain in the butt that scanning the code is dramatically more efficient than the alternative. In other words, QR codes need real use cases and I think I found a pretty good one.

Electronic bill payment is too hard

Here’s a bill I just got from my garbage collector resource recovery service. This bill appears in my mailbox only 4 times a year, so I’ve never set up e-payments. It’s just too much hassle. But I recently decided to give it a go…

Step 1

Parse the bill to figure out how much I owe and how to pay online. Note that the URL for bill payment is on the return envelope, not on the bill.

Garbage bill

Steps 2-21

Now it’s time to set up e-pay! I counted a minimum of 20 steps in this process, not including the confirmation page.

e-Bill Pay Setup is over 20 steps

That sucked. How can we make it better?

The process I outlined above is typical for setting up e-bill payments and a typical person might set up well over a dozen of these. What if we could simplify it using QR codes?

During a one-time setup process, you link your credit card or bank account to a centralized bill pay service and install an app on your phone. Next, simply open a bill, any bill, find the QR code and scan. The bill is automatically paid after a simple confirmation of the payment amount. That’s it.

Pay Now appears at the bottom of all of your bills.

Here’s what it might look like on my garbage bill:

Garbage bill with QR code payment

As you can see, this is a dramatically simpler process than setting up e-billing for each of your accounts. It easily eclipses the cost of opening the app and scanning the code.

The benefits could extend beyond one-time payments. After scanning a code, the app could ask you if you want to eliminate statements by mail, if you want to set up recurring payments, etc. If your credit card expiry date changes or you switch banks, you can update your info in one place for all billers.

Out of the hands of marketers, into the hands of designers

QR codes may never take off, but if there’s any chance for them they need to be moved out of the hands of marketers and into the hands of designers. We need to find use cases of real value, like removing 20 steps from a bill pay process, not linking you to a promotional web page about Chips Ahoy cookies.

I had seriously considered just paying the balance in full rather than put up with the site, but I trudged through the painful new online payment process and got my account set up for recurring payments. Or so I thought. Today I received a statement in the mail informing me that I am 2 months past due. How could this happen? Again?? Upon closer inspection of the e-payment setup process I discovered the problem: the form only asks for checking account number once. Unlike nearly every other electronic payment form, they didn’t provide two account number fields to be checked against each other for accuracy.

My payments never went through because I had transposed 2 numbers in the 10 digit string that is my account number. As a result, my account is past due and I now owe $25 in late fees. Could I have been more careful entering my number? Absolutely. But it’s common for users to enter 10 digit account numbers incorrectly. The form should account for this. When asking for a long, complex number like checking account, always have the user type it in twice, the same as they would for a password. That way, transposed numbers are easily discovered, you get paid, and your customers don’t flame you on the Web.

If you want to get a sense of how complex it is to set up a fairly typical TV configuration, check out the play-by-play of my recent experience switching cable companies below. This was going to be my original post, but it’s about as tedious to read as it was to experience. Still, if you’re interested in how broken the TV experience is, read on…

The ordeal that is cable TV

Last week I got fed up with Comcast and decided to switch to another provider (Astound, if you’re curious). I was essentially getting the same services from the new provider as I got from Comcast, so this should have been a pretty straightforward process.  The experience I will outline below is fairly typical and, in my mind, a fantastic example of why the world needs an honest to goodness TV from Apple.

Cable guy

“Hi, I’m on my way. I got a cable card for your Tivo. To be honest, I don’t now much about these, so I hope you can help me out.”

Me

“I’ve set one up before when the last cable guy didn’t know how. I kinda remember the process.”

Cable guy

“Whew! That’s a relief! See you in a bit.”

I hop on the Web and search for Tivo + Cable Card on Astound. Lots of complaining about how hard it is to set up, how it never got set up, etc. Starting to look grim.

Cable guy

“Knock, knock! It’s the Cable Guy! Let’s set up your Internet first.”

Within 5 minutes, Internet is up and running. All I had to do was unplug and re-plug the power on my AirPort Extreme. Done (and it’s wicked fast!).

Cable guy

“I love this Apple stuff. Never have any problems. Now, here’s the cable card for your Tivo. Show me how this works.”

I plug in the cable card and see this:

Cable guy

“That’s funny. Maybe the card isn’t activated. Let me call the service desk.”

Several minutes later, they send a “signal” to the card. We check to see if we get channels. Success!

Cable guy

“I guess that’s it!”

Me

“Yep, I guess so. Thanks!”

I say my goodbyes to the Cable Guy and he drives off. I start using the TV and realize that none of the channels match the guide. Some channels I’m supposed to receive aren’t there, the ones that are there have the wrong names, other channels just come up blank. I decide to call support to troubleshoot.

Support Guy

“I’m sending a signal to your cable card… Did that fix it?”

Me

“No. No it did not.”

Support Guy

“Okay, I’ll send the tech back out to fix this.”

Me

“Oh you mean the guy who doesn’t know anything about Tivo and cable cards? Good idea.”

I head to the Internet to find answers. After searching and clicking and digging for about 10 minutes I find what I’m looking for. Turns out I have to tell the Tivo which cable service I’m using. After giving Tivo the details, I’m presented with this:

It’s downloading program information for about 30 minutes using a 50mb Internet connection. It spends 90% of this time “preparing” — whatever that means.

Okay, now the lineup matches but I have dozens of non-HD channels, shopping channels, foreign language channels, public access channels and empty channels that I don’t want to see. Surely there must be a simple way to get rid of these?

Another search on the Tivo website tells me I can get rid of channels I don’t want, but there’s a catch. I have to remove them ONE-BY-ONE and I can’t see the channel content while removing channels, so I have to continually hop back and forth between watching TV and drilling into the menu to remove channels.

Removing channels from Tivo is a pain in the ass

At this point, I’ve been setting up my TV service for about 4 hours and it’s still not right but it’s as close as I have the patience to endure. I’m done for now.

Now imagine your average consumer. They would almost certainly have given up as soon as they heard the words “cable card”. It shouldn’t be this hard. How should it work?

• Plug the TV into the wall
• Plug the coax cable into the back of the TV
• Plug the Ethernet cable into the back of the TV (or configure wireless)
• Turn on the TV and enjoy

I should not have to see standard definition channels when I get those same channels in HD. I should be able to remove channels while I’m watching them, and not have to drill into a menu. I should be able to favorite channels while I’m watching them. I should be able to configure all of my streaming services from a computer, smartphone or tablet so that I don’t have to use the stupid on-screen keyboard on the TV.

Have you seen the new Path? Isn’t it awesome!? Why can’t we do something like that??

And who can blame you? Path is awesome! You should make something like it. But don’t kid yourself. The Path app is as rare as a magical unicorn, delivered from the heavens on a double rainbow.

Path feels hand crafted. The app absolutely oozes quality and an obsession with detail. The parallax background scrolling, the detailed textures, the cute little clock that shows where you are in the timeline. They are all delightful. But there is no better example of this than the “new” menu. That simple little plus button that sits nicely in thumb’s reach. Tap it and it responds immediately, with a springy little animation that says, “let’s do this”! Closing the menu is even more fun as the icons spin and collapse back into the button (see video below).

Releasing a product with this level of quality and attention to detail is incredibly rare. It practically never happens. Why not? Because even when a product team has been very conscious of creating a good experience, the vast majority of teams will eventually sacrifice bits of that experience to ship a product. It takes incredible discipline (and time, and money) to avoid making small sacrifices along the way that will eventually lead to a poor experience or, at the very least, one that doesn’t measure up to the level of the Path app.

For Path, the payoff for their attention to detail has been big. As of a few days ago, Path users were sharing 12 moments a second. The Path app has gained 30x the number of daily active users they used to have in the two and a half weeks since the app’s release and they have had 1 million users share their wake / sleep cycles (source).

In an interview at this year’s Le Web conference (video at bottom of post) Dave Morin, founder of Path, talks about what went into the design and development of Path 2.0. Here are the highlights:

• Morin and team think of Path 2.0 as a “modern journal” and reached this point of clarity by talking with their customers
• It took 1 year of design and development to deliver the current iPhone and Android apps
• Morin describes Path as “a completely design-driven organization”
• Path chose not to develop their app in HTML5 because they wanted the best possible performance, smooth scrolling, etc.
• Some advice:
• “Simplicity takes time… a lot of people launch too early”
• Morin doesn’t recommend shipping minimum viable products to the app stores because of the high cost in bad reviews.
• “Mobile iterations take a long time, so we spend a lot of extra time on quality”

So before you or your team waste a lot of energy being frustrated about why you can’t make an app as good as Path, remember what went into making that app and ask yourself if your team or organization are willing to apply the discipline, time and money needed to make something as good as Path 2.0. In the vast majority of cases, the answer is no and that’s precisely why an app like Path gets everyone’s attention.

We had recently adopted Agile development and so the question for me was, when are you going to “get to done” on the design? My answer was honest but, at the time, not appreciated. I said:

A design is never done, it only approaches done. *

I knew what the reaction would be — What the hell are you talking about? Something is either done or it isn’t! We’ll never ship! You’re going to ruin everything!!

But I still stand by that statement, even more now than I did then. It is foolish to think that you will get the design right the first time. It is also foolish to think that there won’t be changes to a design once development begins. And on a larger scale, software products are never done unless they are dead. Life is better when you stop kidding yourself and your team and admit that there is no “done”.

As a design manager, I use this concept of approaching done to evaluate the health of a project and the effectiveness of its designer. A designer is on the right path when his or her solution is approaching done. In the early stages of a project, there is typically a high level of variability. Designs may appear to be approaching done only to get completely reset once a new requirement or constraint is encountered. This is completely normal. But over time, this variability should become smaller and smaller. That is when you know a design is on the right path. There is a steady progression toward done.

* This idea of approaching done was inspired by some bastardization of what I remember from calculus class and the limit of a function. It may have come to me in a fever dream.

Recommended Reading

There are some compatible ideas in the Lean UX arena. If you’re interested in learning more, I suggest checking out the following sources:

• 9 Principles of Lean UX
• Getting Out of the Deliverables Business
• Lean UX, Product Stewardship, and Integrated Teams

Would you trust this ATM?

My office is located in downtown San Francisco and the surrounding neighborhood can get a little sketchy. Once in a while my route to the office takes me past this ATM machine. Each time I walk past I wonder, who in their right mind would use this thing? Then I remember where I am and keep walking.

Based on what you see, would you tust this ATM? Probably not. And yet there it stands. My guess is that this ATM machine functions just as well as any other. You may want a friend to act as lookout so you don’t get jumped during the transaction, but it likely dispenses 20s just as well as the next ATM.

This ATM stands at the corner of a very broad intersection with freeway off-ramps and on-ramps cutting in every direction. It’s not the kind of place you want to wander around and there’s a good chance you won’t find another ATM for blocks. No doubt this works in favor of the ATM actually getting used despite it’s appearance. But imagine if there were a nicer-looking ATM right across the street or right next door?  How often would it be used then?

Okay, so what’s the point? Unless your customers are desperate and have no other reasonable alternatives, design matters. A lot.

Unless your customers are desperate and have no other reasonable alternatives, design matters. A lot.

If you’re drunk, hungry and need some cash for tacos, you will probably take your chances and deal with it. The Internet used to be a lot like this. People were desperate for solutions and happy to have a product that worked. Whether it looked nice and behaved well wasn’t often relevant. It got the job done. But today there is no shortage of companies chasing a few central ideas and there isn’t a lot to differentiate most of these products and services at the feature or technology level. That’s why we see so many companies scrambling for design talent.

Right now, there are so many bad experiences out there that being NOT bad can set you apart. But not for long. Not only does design matter, details matter. It’s not just about making your product look pretty. You need to nail the interactions, you need to get all the little things right. Increasingly, you’ll need to be building products to the level of things like Path and Flipboard or you just won’t be able to compete.

Are we in a Golden Age of Design? It would seem so. But to be sure, there are plenty of companies out there with design talent that is not being fully or properly utilized. Plenty of companies hiring designers to “make it look pretty” instead of participating in the product definition and development processes. Plenty of designers frustrated at the inability to get ideas from paper or prototype to production. In order to succeed going forward, designers, product leaders and developers will need to work more closely than ever and focus on delivering high quality design with an attention to detail.

Problem:

SD card is not recognized when inserted into the slot if the computer is already running, but will be recognized if the card is in place on reboot.

Solution:

Turns out this is not a software problem, but a hardware problem. There is a small switch inside the card reader that signals to the computer that a card has been  inserted. This switch can get stuck.

SD Card slot on iMac

The solution is to insert a toothpick or other non-conductive “stick” into the SD card slot to free up the switch. According to posts on the Apple site, the switch is located at the top/back of the card slot. I also poked around the bottom/back of the slot for safe measure (I never actually saw the switch). After “flossing the inside of the slot a couple times, the reader works once again!

Insert toothpick into slot to "unstick" the switch.

The Command Line Interface Returns

Remember the command line interface (CLI)? Before the popularization of the Graphical User Interface (GUI), computers were controlled by issuing text commands. These commands were often highly efficient, but required users to memorize an endless number of precise instructions in order to get work done. Graphical interfaces made computing more approachable to the average person by introducing visual metaphors to replace complex commands. In many respects, Siri represents a return to the command line interface. The difference is that Siri allows users to speak their commands in natural language. The software interprets these commands and performs the work.

Like the original CLI, voice commands can be incredibly efficient. For example, setting an alarm on my iPhone requires over 15 swipes and taps to complete. I have to unlock the phone, enter my passcode, locate the app, launch it, add an alarm, rotate the dials to set the time and press Done. Using Siri, I simply say: “Wake me up at 6:30 tomorrow morning.” Unlike the CLI, I didn’t need to memorize a command, I just spoke naturally. Here’s what it looks like on the phone:

But I don’t need to see it on my phone because Siri tells me what it’s doing by way of a friendly, slightly robotic voice. We can have a conversation and Siri can even use my previous “commands” to add context to what I say next. For example, I can say “Change it to 7:30″ and Siri will know that by “it” I’m referring to the alarm I just set. Your old green screen terminal probably never did anything as cool as that.

I am grateful to have seen Steve present his last keynote at WWDC. While watching the keynote, I realized this day was probably closer than I wanted to believe – that I may indeed be watching the last keynote. Unfortunately, it was.

Like most of my colleagues in technology and design, I owe a huge debt of gratitude to Steve Jobs and Apple. Were it not for Steve, I wouldn’t have the incredible, rewarding and enjoyable career I have today. As is often said, knowing the end is coming doesn’t make it any easier and this sure ain’t easy. Thank you, Steve.

Kindle Fire Revealed

Amazon announced the Kindle Fire, a new Android-based 7 inch tablet that will be sold for an impressive $199. It’s aimed at making it incredibly easy to consume media and purchase products, both physical and virtual, from Amazon. It could be said that this is the next generation of Amazon shopping cart technology. All indications are that this is just the beginning for Amazon and there has been no shortage of speculation as to what Amazon will do next, from buying WebOS from HP to snapping up Netflix’s streaming business. One thing is for sure, Amazon is suddenly right smack in the middle of the device business with the potential to become a really big player.

iOS Dominates Web Traffic

Despite the fact that there are more Android devices out in the world than iOS devices, iOS continues to dominate Web traffic for mobile devices. According to data from Net Applications, iOS represented  54.65% of Web traffic to their client’s sites versus Android with a mere 16.26% of Web traffic. Why the discrepancy? It’s not entirely clear, but there seems to be mounting evidence that iOS owners use their devices for the Web and apps much more than Android users. Certainly there are a lot of low cost Android devices out there and those devices may not be connected to the Internet. They may instead be used as touch screen feature phones in many cases.

How Many Mobile Platforms? Three More Than There Were Before

The next big trend in mobile seems to be the proliferation of more platforms. Last week we saw the announcement of Tizen, a partnership between Intel and Samsung, and Meltemi, a feature phone OS from Nokia. Meltemi and Tizen are aimed at touch screen devices. It seems even the low-end of the market is looking for touch. Samsung has also been reviving it’s Bada OS and HTC is also rumored to be shopping for it’s own OS. Add to this the fact that Amazon just introduced what amounts to a new platform with Silk, the hybrid browser that will ship with the new Kindle Fire.

If you thought there was fragmentation before, you ain’t seen nothing yet. Hopefully, as we see a proliferation of mobile platforms, we will also see the vast majority of those platforms settle on WebKit as their browser of choice. In any case, this certainly lends a lot of credence to the choice to build mobile Web apps. How else can any one company expect to support so many platforms (see Future Friendly)?