The primary benefit of the NFC ecosystem is consumer peace of mind, in terms of security. NFC payments are the most secure form of mobile transacting, requiring users to hold their phones very close to the reader in order to make a payment (see Figure 2-5). It uses powerful bank-grade security constructs, such as the smart chip on the device (which is nearly impossible for hackers to crack compared with apps that store the consumer’s card data at the OS level or in a cloud service) plus secure over-the-air data connections, multifactor authentication,^[39] and PIN codes to bar any malicious use of the app. If a user loses her phone, one call to the bank or the service provider sets the ecosystem in motion to either block a stolen phone from making unauthorized payments, or remove the payment functionality entirely.

Figure 2-5. Tapping an NFC reader with the Vodafone SmartPass; in 2012, the Vodafone SmartPass pilot in Australia gave ANZ bank customers Galaxy SIII phones with an NFC wallet app preinstalled, to use at the 100,000 shops with contactless terminals in the country

Likewise, studies have shown that consumers are more likely to trust a mobile payment application if it is associated with their financial institution. In the chart shown in Figure 2-6, you can see that financial institutions like card networks and banks clearly win out over tech companies like Google or Apple and mobile network operators like Verizon or AT&T.

Figure 2-6. A survey of consumers by Wedbush Securities in 2013 found that 27% would prefer mobile wallets sponsored by their bank

Secondly, the speed of an NFC transaction itself makes dealing with cash look resolutely Stone Age; some apps allow you tap and pay without opening the app at all (usually for low-value transactions or intransit scenarios). The act of tapping an NFC device feels more natural to some users, as opposed to trying to properly align a barcode or a QR code to a scanner (which has to be an optical scanner, as the lasers of more common scanners will just bounce off the phone’s glass screen). While it is true that adoption lags in the US compared with Canada and Brazil (which have higher penetration of NFC POS readers in retail), NFC can certainly be faster than credit cards and cash. An NFC transaction from start to finish is about a third faster than other forms of payment, as you can see in Table 2-2. The overall time it takes to pull out a leather wallet to remove a credit card, swipe it, type in a PIN, and collect a receipt from the cashier can be around 45 seconds, give or take (go ahead—time yourself!). An NFC payment, if you already have your phone out (which you probably did while you were bored and waiting in line) generally takes around 12 seconds. The additional second or two for the cashier to give you a receipt is also stripped away, as the app can now give you a fairly complete transaction record.

Other benefits are the ability to pay even when there is no data network (like inside a big-box retail store where phone signals are diminished) because the system is built upon existing settlement and processing rails that we use every day with plastic cards. The system’s interoperability with loyalty cards, transit ticketing, government identification, and building access/keyed entry makes it the primary contender for replacing all the uses of the old leather wallet.

Critics of NFC question whether the technology will ever reach wide use, citing the lack of merchant acceptance, and that it doesn’t truly save time at the register when compared with using a debit card or cash. Unfortunately, all the planets of the NFC ecosystem must align before the consumer can download the app and find a shop with a compatible reader. An initiative like Isis^[40]—a rare joint venture of the three largest mobile carriers in the US (AT&T, Verizon, and T-Mobile) whose participants include card issuers like Wells Fargo and American Express—takes an immense amount of coordination and relationship building, which can be a lengthy investment of time. Countries like South Korea and Japan have had success with this system, because it was generally lorded over by individual mobile networks or device manufacturers. Consortiums like these were a key element to getting this form of payment into the hands of consumers. The nuanced interplay of these parties has a noticeable influence on consumer experience, especially during the first use, when the user is setting up his phone for payments and downloading his card data.

For example, with some NFC apps, after the user downloads the app and registers, it may take anywhere from 30 seconds to 12 hours until the bank authorizes the TSM to install the user’s card on the device. Designers would have to consider how to communicate to users that they must wait for a significant amount of time until their phones are ready to tap. Communication during the activation process must be transparent and reassuring, without overwhelming users with technical jargon. Any ambiguity during the onboarding process tends to make a bad first impression.

Another challenge is feedback in the form of information that can be shown to the user after the transaction is complete—essentially taking the place of a receipt. In some markets, the contactless readers at the register may not provide the app with complete transaction data when users tap their phones, such as the purchase amount or currency; they may send back only a vague “transaction data received” response, which can be somewhat confusing to consumers, causing them to rely on more traditional stimuli: a nod from the cashier and a beep from the POS to tell them the transaction was approved.

Merchant acceptance is always a challenge for mobile payment apps, and NFC is no exception (though it is the most widely accepted). The number of NFC-enabled points of sale varies from region to region. Places like Canada, France, Brazil, and Poland have contactless readers in every corner shop and taxi, whereas as of June 2013 there were only 500,000+ readers in the US. This number may grow within the next few years, thanks to the payment networks (Visa, MasterCard, American Express, etc.) adjusting their risk liability policies to place more responsibility on the merchants for fraudulent purchases, which would be reduced by EMV cards like the ones used for NFC. The burden will soon be on merchants to make sure their POS systems are upgraded with the latest accessories. But for now, just over 7% of the 7 million retail stores in the US accept NFC.

Another frustrating challenge to implementing an NFC payment app, and one that designers and developers typically have little to no influence over, is overall awareness. This does not apply just to potential users being aware of the service and its benefits (which is why we hire marketing folks). Consumer awareness is relatively easy to get a handle on, as more NFC-enabled devices are released into the market and users become more accustomed to using NFC to share pictures or check in at their favorite venues. A clear, strong marketing campaign combined with a successful user experience will help uneasy users feel more comfortable using their phones in this way.

But once the reader is up and running, a store may or may not have properly trained its staff in how it all works, prompting blank looks from the cashier if the customer has any trouble tapping her phone. Sometimes this kind of training is just not a priority, given how few NFC users are out there relative to people paying with cash or cards. More than a few times, I’ve pulled out my Google Wallet or Isis Wallet on my NFC-enabled Android phone at my local Whole Foods, only to find the employees had helpfully covered up their contactless reader pads with a placard plugging the weekly sales promotion or marketing (see Figure 2-7). Even if your favorite coffee shop does have a compatible reader, it must be correctly configured and in good working condition to accept a tap (from any contactless form factor, be it a phone or a card or a keyfob).

Figure 2-7. An NFC reader at Whole Foods, partially obscured by a placard

Cloud wallets are an agile alternative to the complex infrastructure of NFC, with much lower barriers to entry for both users and developers. A new startup looking to enter this space need only comply with cardholder privacy rules set by the Payment Card Industry Data Security Standard (PCI DSS), established by Visa, MasterCard, American Express, JCB, and Discover to provide multiple layers of protection for handling cardholder data and conducting digital transactions. After PCI compliance and the establishment of a settlement API to process payments, it’s simply a matter of creating a user experience that will authenticate that the user: a) is who he says he is, b) has a valid card, and c) intends to initiate a payment. Registration for a new user is fairly pain-free (once he gets past the dreaded credit card form), and he can be up and running in less time than with NFC apps.

These kinds of apps are also hardware agnostic. There’s no need for users to upgrade their devices to one that is NFC ready, for example, so this approach tends to work on all the major mobile platforms (unless it involves the use of Bluetooth Low Energy beacons, which require phones to meet the latest Bluetooth specifications).

This approach still faces the same distribution and adoption challenges that NFC wallets do, in that both rely on the merchant having compatible hardware at the register—whether it’s a laser scanner for barcodes, or a point-of-sale app that recognizes when registered customers are nearby. When it comes to revenue, merchants tend to stick with what works and aren’t often prone to experimenting with register hardware—especially if they just dropped $5,000 on a point-of-sale system.

Another hit to cloud payments is that they’re not as favorable to merchants who must work with an acquirer to process their transactions, because these acquirers typically charge higher processing fees for transactions perceived as being higher risk. Usually cloud-based payments are considered “card-not-present transactions” for which the store will incur higher fees.^[42] This is in contrast to NFC payments, which act as “card-present transactions,” meaning the cardholder and the card were present at the sale, so they carry a low settlement risk.

Cloud payment apps expose the user to the most security risk, once her card data has been transported and stored on the server. Though most apps like PayPal’s allow the user to set a PIN or passcode to lock her app, the app is only as secure as the cloud server, the encryption strength of the app’s communication, and the user’s PIN selections. In a recent survey, nearly 9,000 iPhone users admitted that their screen lock code is 1234.^[43] The second most popular lock code was 0000! Should a hacker eavesdrop on a user’s mobile data traffic, using methods like Wi-Fi network spoofing—that is, setting up an innocuous-looking public Wi-Fi network called “attwifi-free” or “free public Wi-Fi”—then any phone that hops onto this network (either by user choice or because the network resembles one the phone already trusts) provides a window for that hacker to see passwords and cardholder information being passed between the app and the cloud. That double latte might cost more than the user bargained for!

Another practical challenge here is that transactions aren’t possible when the cellular network is unreliable or Wi-Fi is unavailable (see Figure 2-12). Even when data connectivity is not an issue, there is often a perceived latency while the user waits for confirmation of the transaction in the form of a push notification, SMS, or some other form of digital receipt. Even in today’s world of lightning-fast 4G LTE networks and public Wi-Fi hotspots, just a short connectivity drop of a few bars is enough to slow or halt a cloud-based payment.

Figure 2-12. Cloud apps depend on data connectivity to facilitate payments in person; lapses in data connection often render them useless

Closed loop apps are inherently easy to implement from an acceptance perspective, as they are usually sponsored by the merchant itself. They are especially quick to release, thanks to the use of barcodes and QR codes, which can be read by most laser scanners already attached to modern points of sale (though it’s unlikely that you’ll find them at the corner grocery store). All the checkout cashier needs to do to accept the payment is to indicate that the customer is paying with a gift card.

Figure 2-13. Dunkin Donuts lets users store and reload their gift cards in an app, which tracks purchase history and rewards (courtesy of Dunkin Donuts)

The user’s risk can be very low, depending on how much money he loads onto the card. These low stakes make this system a good entry point for users who want to try out mobile payments for the first time. As these cards are often tied to rewards programs, the user can also be shown a points tracker, which encourages him to earn enough points to get free items and save money on future purchases. Because it’s on a mobile device, a closed loop app that can track the balance of a store’s card is a huge improvement over the plastic version; the user won’t have to ask the merchant to swipe the card to look up the balance, or be surprised when the balance is lower than he expected. These apps are entirely device agnostic, and can be deployed on any mobile device...or even smartwatches (Figure 2-14)! They can be used even when there isn’t a data signal—a leg up on cloud-based payments.

Figure 2-14. The author, paying with a Pebble smartwatch, linked to a Starbucks app

These experiences do need wireless connectivity to an extent, so that the user can see her current balance before she attempts to scan it at the register (lest she incur angry stares from her fellow shoppers for slowing down the checkout line). Otherwise, she is going into the purchase without knowing if she has enough funds.

Closed loops systems are just that, closed: they can be used only at one associated retailer. This clutters the user’s mobile home screen with icons of her favorite shops, as opposed to offering her “one wallet to pay them all.” There are gift card wallets, like Gyft and GoWallet, that can store cards from multiple merchants, but they won’t include rewards points and may not have balance tracking.

A minor UX challenge here is the brightness of the user’s screen and ambient light reflection. Barcode scanners can be fickle creatures, needing just the right light conditions and laser alignment for a good read. Starbucks has utilized the mobile OS to reduce scanning issues by overriding the device’s brightness setting so that the screen shines at the brightest level when the barcode is displayed.

QR codes are not standardized, so they’re vulnerable to malicious redirection or copying by scammers. 1D barcodes are standardized but are limited to 10–20 characters, which doesn’t allow them to store much data, like randomized tokens. They are both easily duplicated. If the user leaves her phone unattended while the code is displaying, someone could (in theory) grab the phone, take a screenshot, and email it to himself.