Wednesday, December 25, 2019

A glitch in the payments at the U2 concert, and lessons for design principles

by Sanjay Jain, Rajeswari Sengupta, Ajay Shah.

On 15 December 2019, for the first time, U2 was to perform in Bombay. The organisers of the concert came up with an elegant vision for how 50,000 people would be fed: This would be done through an all-electronic payments process.

It was supposed to be all pretty and perfect. Along with the tickets, everyone got a card that contained an RFID tag. The card had to be activated by scanning the inbuilt QR code using the android QR code app. This took customers to a URL. On the website customers were required to register (with a name, phone number etc) and then pre-load the card. It was announced that the pre-loading could be done exclusively via a particular payments app. Once this step was completed, customers would need to tap their cards at any of the physical kiosks at the venue in order to update the card with the online balance.

Front

Back

Customers were informed that food and beverages could only be purchased at the concert venue using the RFID card, no refunds would be available and that there would be physical kiosks at the venue for topping up the cards. While those counters would accept both cash and cards for payment, the food and beverage sellers would only accept the RFID card.

Thus, a key part of the design vision was coercion of the customer by forcing all payments for purchases to be done only through one mode.

Failures in consumer protection


The cashless monopoly electronic payments mechanism had unhappy features from the viewpoint of a consumer. The design was complex, requiring customers to first do an online transaction for loading the RFID card and then a physical transaction at the venue for getting the amount to the card, even before the card could be used for purchases. In other words, two additional steps were required to potentially save time during a future purchase transaction.

The minimum recharge amount for the card was Rs.500. Customers were forced to do this without any upfront knowledge about the prices of the goods sold at the venue.

To load Rs.1000 into the wallet, the customer was charged a fee of Rs.100. On this fee, there was a GST of 18%, so the payment went up to Rs.1118. This was a total cost to the consumer of 11.8%, as compared with paying cash.

To add insult to injury, the money left on the card was not refundable. What was not spent would be confiscated by the payments vendor. Customers were thus required to estimate their expenses without knowing the prices of the goods up front. This disrupted the biggest benefit of digital payments i.e. the ability to make a purchase without planning for it beforehand.

Even in the best case scenario where the design envisioned by the organisers worked, customers were left with an experience worse than a cash payment, for a significantly higher price. Even in its conception, this was not a very attractive grand scheme.

Failures in system operation


In the event, the grand scheme collapsed because it just did not work. There rapidly emerged two classes of customers at the venue.

One class was represented by customer X who had pre-loaded the RFID card with Rs.500 using the app before arriving at the concert venue. At the venue she found that her F&B spend would be (say) Rs 1000. Despite having taken the trouble of registering and pre-loading the card prior to the concert, she now had to stand in two long queues at two separate kiosks: one to first update the card so that the online transaction of Rs 500 was fed into the card, and the other to top-up the card with Rs 500 so that she could make her F&B purchases.

There were thousands of such customers who went from one queue to another even before their registered cards could be used. The queues at the payment counters soon became longer than the queues at the F&B counters.

The other class was represented by customer Y who had not registered or activated her card before the concert. At the venue she had to first activate her RFID card using mobile data connectivity, load the minimum amount of Rs.500 online, and then stand in the queue to update the card so that she could use it for purchases. In case she wanted to top-up the card, she would have to stand in the queue again.

Given the large fees (11.8% plus the possibility of non-refund), users were careful to avoid putting too much money into the card, and thus ran the risk of undershooting.

At first the hassles were only about standing in multiple queues. Things got worse when the systems at the kiosks for updating the registered RFID cards stopped working. This meant that thousands of customers who had pre-loaded their cards could neither use their cards to make purchases nor get refunds.

The venue was swamped with a large number of queues for one thing or another.

The mobile data network crashed at the venue with thousands of people trying to access it. This meant that thousands of customers who had not pre-activated their cards, were unable to do so, let alone update the same and use for purchases.

Despite the utter chaos, the merchants inside the stadium stood firm, in unison, in refusing to sell goods to the customers using any other payment mechanism.

The net result was that thousands of people were left without food and beverages despite possessing all reasonable modes of payment and despite there being multiple merchants selling the desired items. Money -- the bridge between buyers and sellers -- broke down due to the framework of coercion around only one technical standard that was permissible.

This was a mess-up at multiple levels:

  • Poor product experience, requiring multiple redundancies such as to transfer from online payment to the RFID tag,
  • Poor planning on the part of the organisers and not factoring in the possibility of collapse of data network at a venue filled with thousands of data users,
  • Fleecing of the customers with a steep 10% load charge and forfeiting unused balances,
  • Coercion of customers by not giving them alternative payment options.

Learning about design principles from this episode


Many an engineer can come up with a grand scheme that sounds nice. For an engineer, it is easy to build, and easy to work with simple monolithic systems that are designed by someone and imposed on everyone. The engineer's job is easier, operating in such a world, than in the messy real world of multiple technologies. However, social systems and the interactions of a large number of people are complex, and the best laid plans of designers are likely to go awry.

We are in favour of innovation, and trying out shiny new ideas. Offline instant digital payments through RFID tags sounds like a great idea, particularly when you anticipate a crowd and poor network conditions. However, innovations can and do fail. The wise path is to never have a single point of failure, to always ensure that customers can access other options, that the failure is not as damaging as it was at the U2 concert. Payment is an enabler, and not the final product, and when the payments system actually hampers transactions, it is a tragedy.

The problems of a single centrally planned solution, inside one stadium, help us in thinking about central planning more generally. Each new innovation must face the market test. It is always better to have organic evolution, where many rival solutions slug it out in the marketplace, with no coercion that helps or hinders any one solution. This will give more robust solutions (no single point of failure), let the market evolve towards numerous solutions that fit numerous work environments (e.g. decentralised data works better when communications systems break down, centralised data has its own advantages for certain situations, etc), and prevent any one vendor from ripping off the consumer. It is good to have competition between multiple technologies, and multiple technology choices that fit the very diverse array of use cases that are seen in India.

We have traditionally extolled the role for cash as a way to protect individual privacy and freedom. We must also respect the remarkable UX of physical cash transactions, and contrast this with the hoops that many digital schemes want to force consumers to jump through. The cashless dystopia of the U2 concert teaches us that cash has one more important function: In a disaster zone where IT infrastructure has broken down, cash is the way to get transactions done. Physical pieces of paper will be important for a long, long time.



Sanjay Jain is at CIIE.CO, IIM Ahmedabad.  Rajeswari Sengupta is a researcher at IGIDR, Bombay. Ajay Shah is a researcher at NIPFP, New Delhi.

4 comments:

  1. I had to Shell out 500 rs
    For glass of water
    Remaining balance of no use
    Very long queues at food joints

    ReplyDelete
  2. Sanjay, you have articulated the situation very well. I doubt the intentions of the planners, the design of it points to greed. Cringe inducing greed. The phone and data was totally dead for the entire duration of the concert.

    ReplyDelete
  3. I hope BookMyShow is aware of the mess they created with the F&B arrangement for the concert. It was suppose to be a special night for many who were eager to watch U2 for the first time in India and BookMyShow still found a way to ruin it. The worst part for me was the chaos that ensued in getting your glass of water. The kiosk staff were clueless and many customers took matters into their own hands and started buying cups for others. This is a great post, one that needs to be documented and read by concert managers.

    ReplyDelete
  4. I like the statement 'Payment is an enabler, not the final product'. Cash will always remain an instant gratifier, especially in the sub-100s. Thanks for sharing the wonderful post.
    Cheers.

    ReplyDelete

Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.