2011.03 | Why Starbucks’ Mobile Payment System Works

Last week Starbucks announced that the mobile payment scheme it has been piloting for some time will be available for all 6,800 Starbucks stores and Target locations across the US.  The solution is not yet in place in Canada.

For the uninitiated, the solution works as follows.  Consumers download the Starbucks Card Mobile App to their mobile phone; be it iPhone or BlackBerry. Customers with a Starbucks stored value card (effectively a gift card) that is registered on the Starbucks website, enter the card number into their phone when the obtain the app, and that card number is stored.  When consumers visit a store, they place their coffee order as usual, and indicate their desire to tender with their mobile.  Consumers start the Starbucks Card Mobile App on their mobile and navigate to the payment screen so that a 2d barcode representing the consumers’ Starbucks card is displayed.  The Starbucks associate, selects mobile as the tender in the POS, and prompts the consumer to use the customer facing imager (the same as those used in airports to read boarding passes).  The consumer places their mobile device under the imager, the 2d barcode is read, and the POS treats the tender like a gift card, following the usual payment verification procedure.  Once tender is complete, the customer obtains their coffee as usual.

The discussion on electronic wallet is an industry favourite, and this development will certainly encourage more discussion on the subject and provide some much need experience.  I’m fully behind this initiative, but at present, this solution is very much a Starbucks specific solution, and it is not easily translatable to other retailers.  While retailers can learn a great deal from the obvious careful thought that has gone into the solution, and we can look forward to others moving down this road as well.  To clarify for consumers (and non-technical retail executives) who ask why other retailers don’t have mobile payment schemes as yet, consider the following unique characteristics of the Starbucks situation that make a solution like this pay off.

Use of Stored Value Card – Very few retailers have a stored value card with the massive following and ongoing usage that Starbucks have.  Effectively consumers are giving Starbucks their funds in advance in exchange for some very small benefits (free drink on your birthday, free pump of flavouring in your drink).  Starbucks gets loyalty data on customers, and a nice balance of cash on hand.  More relevant to the mobile payment solution, the Starbucks mobile phone application allows consumers to make a payment onto their stored value card, and the application’s 2d barcode payment system is connected to that card.  Connecting the mobile payment system to the stored value card means that Starbucks can take the risk of a payment system internally.  Stored value are not subject to the same roadblocks, legislation, and scrutiny that building a mobile payment system that would access a credit card or a debit card would have.  Using the stored value card simplifies implementation and sidesteps many complexities of payment systems like EMV and PCI.

Cross Platform – While Starbucks are very keen on the iPhone, they have not limited themselves to an iPhone app, but also provided an app for the other key smart phone users via the Blackberry App.  Considering the corporate core of Blackberry users and how often meetings now take place in Starbucks stores, this is a wise move to maximize potential users.  Given the number of Android Users and the recent release and growing use of Windows 7 Phone platforms, it would not be surprising to see the Starbucks Card App ported to those platforms as well, ensuring maximum potential usage.

Valuable App – With over 400,000 apps on iTunes, retailers need to make their app unique and useful.  Ideally it pulls together the mobile and in store experience in some way.  Starbucks has managed both.  Any successful retailer’s mobile app needs something unique to it to encourage download, and having it on a consumers screen on a permanent basis.

Customer Demographic – Based on my experience, and what I have read in the media over the past few years, the average Starbucks consumer is more likely than average to be a tech-savvy iPhone or Blackberry user, and beyond that, the kind of user who would be comfortable with technology and placing a payment with their mobile.  It is important that any solution put in front of a consumer by a retailer fit their target market.  A savvy comfortable customer is more likely to use the app, and use it well, to speed transactions and drive convenience for them, and speed throughput for the retailer.

Infrastructure – Most Starbucks locations have 2 terminals.  In order to leverage 2d barcodes, special imagers are required, and this means hardware investment.  2 lanes means only a $300-$400 investment per store for imaging hardware.  Considering the potential value of transactions per store, this is a very low cost.   The ROI would be far less attractive for a lower margin retailer with dozens of lanes in a store to deploy, as it would be key to have the imagers in every lane to simplify the process for consumers.

Transaction Type – The slowest portion of any retail transaction, and the most difficult to trim time from, is the tendering process.  Given that in Starbucks transactions generally include a small basket size and the ordering time is relatively short, the value of an alternative payment is increased, as it is a greater proportion of the transaction.   This value is increased further by the incredible traffic at Starbucks sites.  Having many small transactions provides a boost to the ROI of the solution.

No Mobile Device Handling – In order for any sort of mobile payment solution to increase throughput and minimize operational complications, it is key to streamline the process of scanning the mobile device.  Starbucks has done this via a customer facing scanner with very simple signage.  This allows the consumer to place their phone in the scanning area with no need to pass the mobile device to a cashier.  This simplifies the process by providing a consistent process, not only increasing the scan speed, but also avoiding the potential of store staff dropping or otherwise damaging a customer’s mobile device.  Consumers are also more likely to use the mobile payment solution if they do not have to pass their mobile phone to a cashier, given how consumers increasingly consider the mobile device as a personal item.

As with all solutions implemented by consumer facing organizations, ROI is key.  Looking at the Starbucks solution, the costs of entry are probably not that high.  A mobile app is relatively inexpensive and standalone compared to other point of sale solution implementations.  Using the stored value card leverages electronic processes and databases already in place.  The crucial part is operationalizing the solution, and that can be put in place for hundreds or low thousands per site.  All in all, this is a relatively low cost solution with the potential for a high ROI in both funds, and in good will from consumers.  Other retailers looking to implement such a solution would do well to observe what Starbucks have done, but note well that this is not a one size fits all solution.  Any future implementers should be sure that the app suits their customer demographic, their transaction model, and has a way of dealing with the complexities of payment.  Other solutions will arise, and it will be fascinating to see what comes next.


  1. Thanks for the comments and for reading. As always the answer will be that the investment depends on what you have and what you wish to achieve. Starbucks made sense as they had so much of the infrastructure in place, and the solution fit. If that is in place for you, I suggest always starting with your POS software provider to see what they suggest and work from there.

  2. Hi Tim,

    I must say this is one of the best analysis i have seen on 2d barcode payment, you did break it down.Really would love to know the level of investment that has to into developing a 2d barcode payment system and what firm(if you know any) can pull this off

  3. Tim Dickey says:

    Thanks, Brani. I suggest that if a stored value card is a useful mode of payment accepted by a retailer’s clients, than using this method is a good interim step to mobile payments via NFC. If a stored value card isn’t a payment method that is useful and acceptable to customers, I think it’s better to just wait, as building a payment system would be a waste of valuable retailer capital and effort.

  4. Nice work Tim, I enjoyed that your not obscured by the gadgetry of it all. Do you suggest that integrating stored value cards with the mobile for the time being is a pretty good solution at least untill NFC arrives?

  5. Tim Dickey says:

    Hey Jim,

    Thanks for the comments. You pose great questions, and I only wish there were a simple response! They are more like great discussion starters. I expect that a large QSR with many banners would have the same issues with a mobile ordering and payment solution that they would have with their POS and their operations in general. Whatever is implemented has to fit the operations and be optimized to be a part of the front end. Without a simple consistent interface in the stores any solution will be a failure. If it can be made it generic, that would be great from a support and logistical perspective but more importantly it needs to suit the needs of the consumer or it will fail.

    Whatever is built is highly dependent on the operations at the specific site. Your comment about skipping the queue is a great example. There is a benefit to savvy clients of ordering on a mobile app and skipping the line, but while that worked in Lenny’s, my belief is that has the potential to drive bad feelings in a store, and gum up very carefully scripted operations through multiple queues of orders – it might not, but why chance it? What is the upside to the operator? In my opinion, it would be better to build a solution that provides a benefit that can be used by all, but doesn’t unnecessarily penalize anyone. For example, consider adding an ordering capability to the Starbucks app that adds your drink order to the 2d barcode for payment. Instead of sending it ahead of time and trying to figure out which drink is yours (or from Sbux perspective – you not showing up) you scan one code to order and to pay. Now you have streamlined the ordering process, and the barista confirms your order instead of mangling it. This doesn’t penalize anyone in the queue and if it speeds up the whole process, you can expect others to take part. The benefit of this sort of system is mutiplied on more complicated orders. What if your wife could set up an order for dinner online or on her app, and then you just scan the code from your app at the QSR on the way home. This would streamline a queue even more, and those behind us in line would be less likely to bail.

    With respect to payment, I would love to see NFC, and it really is the way to go from a technology perspective, but until the telco-handset manufacturer-carrier-credit processing cabal figure out a way to get their percent out of it, we will be consigned to 2d codes or maybe funkier Microsoft Tags.

    Wordy response, I know. While it may be somewhat complex, I think there are ways of doing it. It’s a matter of will and a validation that there is something in it for the operators and the customers.

  6. Tim,

    Great article. I’ve been thinking about how the Starbucks mobile payment model and consumer experience might translate to the fast-food restaurant industry. On the biggest scale, how would a global conglomerate with multiple brands, 30,000+ international locations (some franchisee owned, others corporate owned), and many different POS systems execute a multi-brand mobile order and payment solution? Would be curious to know if you have thoughts on that. Mobile Wallet in the sky? NFC? Integration with different POS? Seems like it would be super-complex, even unachievable.

    When I first used the Starbucks mobile payment option, I was disappointed to learn that I couldn’t place a mobile order (and skip the line). This would likely drive Baristas like J! absolutely bonkers, and cause fistfights in even the most gentile suburbs with Starbucks outposts. I was content to enjoy the technical wizardry of paying via phone, and leave it at that.

    Lastly, I ordered and paid via Splick-It at Lenny’s sub shop yesterday. Mobile Wallet in the sky solution – no loyalty program. It worked pretty well. I jumped the line, saving about 15 minutes (long line). No fistfights erupted.

  7. Blaine,

    I see you may be set on the Ipad platform. However, my company develops custom touch-scren solutions for retailers on Microsoft platforms. An idea I have for Pharmacies would be leveraging the object resognition ability of Surface. For example, a customer registers for loyalty program at pharmacy at home, online, where they store personal and payment information (maybe even a picture for identification purposes). They are issed a card with a persoanlized tag on it that when placd on the Surface, recognizes them and allows them to bill the transaction to their stored account. You can email me to chat more if you’d like.

  8. Tim Dickey says:

    Thanks for reading and the comment J! Timing is always an issue with payments. I have seen a similar problem with contactless (NFC) on some pinpads. The message may say tap, but you need to wait until one of the notification lights comes on before tapping, or the transaction fails.

  9. J. Call says:

    Good article, Tim! I like your analysis on why this solution works specifically for Starbucks, and how it can potentially lead to more widespread use in the future. From an employee standpoint, mobile payments also have the benefit of being more sanitary to (not) handle: I’m not just talking about the preference over cash that has been who knows where, but a lot of customers have greasy, oily, and even crusty smartphones sometimes, and it saves us the trouble of tactfully cleaning someone’s earwax or makeup off our hands without offending the customer while maintaining our speed of service.

    I’ve also noticed that many customers are very eager to use the app, and we often have to ask them to wait until prompted since a premature scan will lock our screen with an error message. A minor nuisance for us, but it definitely shows the enthusiasm customers have for this cool new toy.

  10. Nice article Tim, you have nailed why this solution is so right for Starbucks.

    Blaine, send me an email with your contact, We are working on a solution that may be of interest to you.

  11. Tim Dickey says:

    Thanks for the comment, Blaine. Hope you find someone to help you build it! I think it might be more challenging in a pharmacy than in the coffee shop environment. Starbucks has smartphone apps, lots of users with loaded stored value cards, constant client visits, a lot of tech savvy smartphone toting types, and only 1 or 2 lanes per site. Pharmacy tends to be a bit different, but bundled in the right way – perhaps with extra loyalty points, it could work.

  12. great article. any body want to help me do the same for pharmacies

  13. Tim Dickey says:

    The Starbucks solution with 2D payments is a perfect fit for the unique Starbucks situation and does not preclude accepting mobile NFC payments (though they would need to update their pinpads to NFC versions in Canada in any case). The 2d barcode payment is not one I would recommend for any other retailer unless they were had the same Starbucks characteristics I outlined in the post. Many Canadian retailers are already equipped with NFC pinpads put into place to with the EMV update, so you are right in that most retailers are better positioned for mobile payments via NFC.

    I understand why you ask the question. The logic would be that if Apple and RIM are so close to releasing NFC units, then this will open the floodgates to consumer adoption of mobile payments.

    My thoughts on contactless payments are well documented on the blog under NFC if you want to pick it from the tag cloud. The problem isn’t the technology, it’s how the payments get processed and who gets paid to do it. See my post 2009.20 | What Mobile Wallet, and the article published today on StoreFront Backtalk. The Credit Card companies, and the processors already get their slice, and mobile carriers have been trying to figure a way to get theirs for years now. Both Canada and US mobile carriers have built companies that they all share in to develop the mobile wallet (Isis in the US, and Enstream in Canada) for example.

    NFC mobile wallets can and should happen (you can already stick an NFC tag on your phone if you like), but sorting out who gets paid how, and how funds will stream through a secure system could take some time. To make a long story short, Starbucks gets more use out of their card, consumers can pay with their phone, and Starbucks avoids this mess altogether to provide mobile payment on their own schedule. I think they have made a good gamble on the fact that they can attract early adopters with relatively very little investment, and by the time mobile payments are mainstream, their system will have paid off.

  14. Hey Tim –
    What are your thoughts on contactless POS systems? The rumor is that the next iPhone and iPad will have the NFC technology installed already. Why would the average retailer install the 2-D barcode scanners instead of installing the payment system? For most retailers it seems like the logical choice.

    Even for Starbucks, i’m sure they will also take NFC payments through iPhones etc. Why roll this out now?



  1. […] about using what’s there today and putting the pieces together in a novel way – just as Starbucks did with mobile payment, and Tesco South Korea did with their Virtual Grocery Store in the subway.  Safeway even offers up […]

  2. […] The company’s CEO Howard Shultz made the comment yesterday during Starbucks’ annual meeting. The mobile payment app was rolled out in January to 6,800 stores and has been tested since 2009. The testing found it was the fastest way for customers to pay.  The app uses a bar code scanner system. […]

  3. […] Why Starbucks Mobile Payment System Works « Retail Technology Trends […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: