Published : 6 months ago, on 10/06/2024

Digital Innovation: Insights on Emerging Ecosystems and Business Models for Commercial Banking

Puneet Chhahira, Head of Product Management and Marketing at Infosys Finacle

The commercial banking sector is undergoing a transformative phase in the wake of the digital revolution, presenting a mix of challenges and opportunities. In this article, Puneet Chhahira, Head of Product Management and Marketing at Infosys Finacle, and David Barton-Grimley Fintech Strategy Director at 11:FS, spoke with Venkat ES, Head Treasury Product APAC, Global Payment Solutions at Bank of America, to discuss the future of commercial banking in the rapidly evolving digital environment.

Puneet: To begin our discussion, how are commercial banking products and services evolving and what are the big shifts you see? What are the implications of developments like real-time payments, and FX volatility on products and services evolution?

Venkat: This is a good question because this is usually the trigger for corporate banking innovation. For example, now that real-time payments are a reality, back-end systems and processes must also reorient towards 24×7, real-time gross settlement. Every other corporate bank is latching on to that, corporate treasuries are building their own systems, and even big ERP players are moving towards this. 

Venkat ES, Head Treasury Product APAC, Global Payment Solutions at Bank of America

Also, real-time information and real-time payments are allowing corporate treasurers to cover FX risks in real-time. Say, there’s a large balance in a currency that a corporate doesn’t want to hold. Because it has real-time information, it can pull the liquidity and convert it to some other currency before the banking system closes. But this is not only about corporates initiating real-time payments; it also applies to receivables. Take an FMCG company whose credit appetite may not extend throughout its distribution chain. With a mechanism like a digital wallet, it can receive payments from even mom and pop stores in real-time, and then trigger the consignment. This is virtually an ecommerce arrangement on the B2B side. It helps to manage risk, and supports credit and sales processes, to result in higher profitability. 

Also, treasurers always seek to effectively manage cash flow, and bring down days sales outstanding in accounts receivables. Without this, it is not possible to do more business through the distribution chain. By enhancing these ratios, the economic value-added indicators are elevated, but at the same time the company is able to service the distributors in a way that both customer experience and loyalty build up. This is where banks and corporates are collaborating to see if they can adopt Artificial Intelligence (AI) or Machine Learning to improve reconciliation, reduce days sales outstanding, and thereby lower the cost of working capital by managing liquidity better. Another area of collaboration is using analytics and AI to study the behavioral pattern of cash flows to make more realistic forecasts than what the treasury systems are able to do at present. 

Beyond this, I believe that technology is transforming things in such a way that in future, everything – like a loan repayment for instance – will be auto-triggered based on an event. 

Puneet: How do you see delivery evolving? The business has always relied on relationship managers but with treasury connectivity and powerful corporate portals how do you see embedded finance evolving in the organizations you serve? 

Venkat: There are large corporates, commercial banking clients and SMEs in a supply chain, and then there is the end consumer. A significant trend across the spectrum is that the bulk of sales is moving from offline to online; take a sector like heavy equipment machinery, now even tractors are sold online. Also, corporate banking was only talking of B2B earlier, that is now changing to B2B2C. This is largely because of the improvement in infrastructure and payment mechanisms, and innovations like BNPL, that make it possible to distribute payments across the payment value chain in a way that financing happens without undue risk. I see traditional trade financing methods getting converged and the industry transforming so there are no impediments and bottlenecks in the supply chain. There will be more efficiency, velocity, sales will increase, and how one manages risk from end to end will be most important. 

David: How is the industry moving towards APIs? Are you seeing any clients go from offline to online to APIs?

Venkat: API adoption is pervasive across industries, from FMCG to insurance to power to branded retail. In Europe, APIs is seen as a bank-agnostic channel that is easier to implement, less expensive than SWIFT (the other bank-agnostic channel), and serves the purpose well as long as security is taken care of. 

In Asia, it is still gaining momentum, as each country is going about adopting standards separately. In North America, APIs are being implemented in a big way. In the earlier example of using real-time information and payments to manage FX, it is APIs that enable the flow of both information and money in real-time. This is extending to the lending side as well. For example, if I have to draw down on my loan, I don’t have to request that using paper or host to host integration. As long as the limits are available, I can do it through API. 

Puneet: Businesses are seeking to integrate banking services across on different enterprise systems – like accounting platforms, ERP or treasury management systems. Is that a challenge when delivering products? And how should banks manage that?

Venkat: The challenge is legacy infrastructure, both on the bank side and also the different versions of legacy ERP platforms being used by different corporates. What’s worse, these are layered on top with other solutions, such as treasury and other products that evolve over time. When you want to deploy an API, it is directed at a particular system, predominantly the treasury management system to initiate a payment or consolidate liquidity; so, you encounter challenges based on the technology situation of the corporate. The thing is, no one wants to make too many changes, and everyone wants an expedient solution. Unfortunately, this does not address the core problem. Simplification of architecture and processes is the need of the hour and every organization must provide the resources and a roadmap for this so they can be future-proof. But not many can afford it, which leaves most organizations focusing on tactical rather than strategic things. This is where banks come into the picture, on our own or along with a technology partner, to offer a solution based on the current stage of the organization, its needs, how and where it plans to consume the API to fulfil that need and we provide a solution accordingly. 

Puneet: How is the corporate banking world adopting machine learning algorithms? Please share your view on the current state of AI adoption and its future. 

Venkat: AI is a reality, and it is fast evolving. Everyone wants efficiency and better solutions. They are using AI for different solutions, not just forecasting. I will give you an example of how Bank of America has used AI/ML successfully to manage receivables. Clients get lots of credits into their accounts and even today, because of the legacy clearing and settlement mechanisms, they are unable to identify who has paid, and for what. It is even more complicated if an invoice is not paid in full.  Sure, ERP and treasury systems are codified, but they are very inefficient at reconciliation. So, Bank of America’s digital banking platform CashPro has brought AI/ML to solve this and using historical data analysis and intelligence – plus rule-based logic – which can identify the source of the payment. Rule-based logic is then applied to understand if a discount or dispute or some other reasons is behind the short payment. After reconciling this, the credit is applied automatically. This is much more efficient. On our CashPro platform, we observed that for corporates with 200,000 or more invoices, match rate went up from 40% to 85-90% within 6-9 months.  

There can be many such instances of AI deployment; with generative AI it is even more powerful. You will see many more use cases in the days to come; like APIs, AI is proliferating and getting adopted across industries and client segments.