• Milos Dunjic, AVP, Payments Innovation Technology Solution at TD Bank Group


• 30.10.2017 04:15 am
Innovation , low value payments , Milos Dunjic is technology executive fascinated with and focused on payments and digital innovation. All opinions are his own and not of his employers

There is a fierce debate out there, about what’s better – subscription or pay-per-use/view model of services? Digital content providers, telcos and TV broadcasters mainly prefer subscription or licensing model as it allows them to better predict revenue than in a pay-per-use/view schemes. But is it necessarily beneficial and fairer for the consumer? Many of us are reluctant to commit and may prefer metered usage for product or service. In my opinion, pay-per-use/view is certainly much fairer and more competitive model than a flat monthly subscription fee. The pay-per-use works best when the usage of the service can be effectively measured. News articles, streaming services, TV broadcasts, wireless plans, etc., are good examples. Still, in most of these cases, the subscription models dominate. Why can’t we read a single news or scientific article and pay only for it? Why can’t we pay for viewing a just a single live match of football on TV? Proponents of the flat subscription fee model argue that pay-per-use model would be:

1. Complex to track, since companies struggle with counting named users reliably
2. Potentially expensive for the clients, arguing that per second/minute/hour/megabyte/gigabyte rate will always cost consumers more than the flat monthly subscription fee
3. Counter-productive, as service providers do not want to create obstacles or disincentives for customers

I am not sure I see it the same way. Amazon already revolutionized the IaaS provider space by introducing flexible and very competitive pay-per-use AWS infrastructure model. If they figured out how to reliably track named users, why can’t providers of digital news and digital streaming? Pay-per-usage model is ideal for infrequent, occasional users, which on average, would spend less than paying for the full monthly or yearly subscription. In my opinion, it is always a good business practice to try and get some revenue from customers than none. I also believe that pay-per-use is not a dis-incentivizing model for the customers, especially if done properly. On the contrary, it may be a great incentive to get new customers on board, instead of pushing them to pay a full subscription and likely losing them forever. Of course, the best would be to have both options available to the consumers and let them flexibly decide how they want to pay for the service usage.

Payment Card Rails Requirements for Pay-Per-Use Model

The current levels of Visa and Mastercard interchange fees may still be somewhat prohibitive for low value payments, which could usually be charged for individual digital news articles, songs, per-minute watching of TV programs, usage of IoT devices, etc. An alternative may be to use cheaper ACH payments, but unfortunately, those are not real-time in nature. Even with the recent ‘same day ACH’ improvements, participating Receiving Depository Financial Institutions (RDFIs) don’t seem to be providing real-time payment authorization for pull requests, nor are participating Originator Depository Financial Institutions (ODFIs) providing an option to push funds in real-time from the consumers to the service providers. What about the existing bitcoin and other public non-permissioned blockchain cryptocurrencies as an option? They could be used eventually, but are still too slow and low-throughput option in my opinion, which urgently requires fresh and innovative re-engineering to become a real-time payments contender. Clearly, the brand-new payment rails are required for cost-efficient processing of individual low-value payments and enablement of pay-per-use implementation models. Such rails must satisfy these requirements:

• offer real-time payment authorizations for pull type payments OR real-time push payments, with consumer’s account debited immediately and payment fully guaranteed to the service providers
• ideally have built-in real-time clearing of transaction totals between counterparties, as opposed to the reliance on the heavy end of day transaction reconciliation and relatively expensive exception processing, prevalent in current payment card and ACH rails
• provide reliable authentication of transacting parties, i.e., be based on modern digital identity and biometric schemes

We may not have such payment alternative available yet, but great news is that the W3C Web Payments Framework, consensus-lite DLTs and Digital Identity schemes are maturing and converging rapidly, enabling payment service providers and FIs to innovate and eventually offer cost effective and guaranteed payment options outside of current legacy payment channels. Hopefully, soon enough we may indeed be able to seamlessly and cost-effectively pay-per-view individual digital content items or pay-per-use of the connected IoT devices.