Buyers' brief: The future of bank treasury tech is AI

By Chrissy Chiu

7 December 2020

Long the bastion of cash and liquidity management, bank treasury teams do far more than position-keeping. Real-time risk management, operations, payments, and financial risk management are rolled into the business function. Over the past few years, market-leading software has become key to overseeing the risk and reward of each process, and so procuring the right platform is essential.

Buying or adding to software is no longer about traditional deposits, tax, corporate and consumer payments. It’s about how treasury teams make connections, and lead the management of new products, technologies and services. Treasury management software needs to be flexible enough to provide treasury teams with new tech opportunities while still delivering core functions.

“One element of the treasury management function is connectivity. Every bank provides access via online banking system and as corporates usually work with multiple banks, they prefer bank agnostic connectivity,” says Ad van der Poel, co-head of product management, global transaction services, EMEA at Bank of America.

To connect with more payment counterparties, treasurers are looking for advanced technology over wire payments. 

“They [corporates] traditionally use SWIFT, but with other tech providers and the development of Open Banking and APIs, we notice that more treasurers are asking for API technology,” says van der Poel.

Banks are on the forefront of innovating and developing new technologies and APIs. JP Morgan’s treasury APIs are already being marketed to corporate clients. The bespoke technology helps clients access the bank’s real-time data, and provides visibility across the lifecycle of treasury payments. JP Morgan is not alone – Singaporean bank DBS deploys proprietary API RAPID to increase payment settlement times and operational efficiencies.

Single APIs will soon likely be replaced with an ecosystem of APIs. Termed the ‘third wave’, in Finastra’s White Paper “Payments and API Banking,” the firm expects there to be evolution from ‘solutions’ to platforms. This means that APIs don’t just solve for connectivity issues. Instead, third wave platforms integrate payment initiation, aggregate accounts, and provide services 24/7. Looking ahead, bank treasury teams might encounter an entire ecosystem of APIs providing add-value services.

AI has the potential to make big waves in treasury technology. AI could reduce the grunt work for treasuries, freeing up time to make strategic decisions. “Another dimension where AI and machine learning helps, with proper human oversight, is risk management. This is particularly relevant where there is a lot of data that requires monitoring,” says Jacqui Kirk, co-head of Product Management, Global Transaction Services, EMEA at Bank of America.

According to McKinsey & Company, AI is already enhancing banking operations by providing real-time transaction monitoring and fraud prevention. Banks are also looking to transform the time-consuming reconciliations process with AI. Asian Banking and Finance magazine says that AI could make quick work out of reconciliations and some banks are already working on how cash management, fees and expense processes can optimised. 

In the UK, the Bank of England’s “Machine learning in UK financial services” report identified AML, risk management, and regulation as areas in which banks are deploying AI as part of their operational frameworks. Capable of ingesting huge amounts of data and learning from human decisions, AI can automate banking treasury operations even further. The technology could potentially help treasurers detect anomalous payments by flagging unusual movements to human controllers. Complex regulation could also be ingested and interpreted, making it easier for treasury teams to comply with new rules.  


In a report by EY, treasury digital transformation goes hand in hand with client demands. Corporate clients will likely expect better technology from banking counterparts. Treasury teams need to be able to provide access to real-time payment information from APIs; increased cash visibility; and demonstrate the value-added service they provide from an advisory perspective.

Functionality to expect from treasury software providers
Van der Poel’s comments haven’t been lost on software providers.

“Treasurers of multi-entity companies prefer a shared service centre for payments and reconciliation of these entities. The ERP system is generally considered as their system of records. Their treasury management station (TMS) is where financials flows are linked to the system. And from there they of course, prefer bank agnostic connectivity to multiple banks [for payments],” says van der Poel.

Murex is one of the world’s leading treasury management software providers. Murex’s MX.3 centralises cash and book exposures across multiple trading platforms. For bank treasurers, the centralisation of information means that decisions can be made for different corporate entities at a single desk.

To keep the treasury desk running, cash control is essential. When choosing the ideal system, treasury teams need to have strict control and oversight of payments and cash.

“And as a treasurer you want accurate cash flow forecasting, as your CFO and senior management will always ask about the company’s cash flow forecasting. Within that structure, you can use a virtual account structure. These virtual accounts are linked to a physical account, and every entity or type of payment flow in the corporate structure would have a virtual account, so you have complete control and data insight,” says van der Poel.

Many treasury management software platforms offer cash flow centralisation. For example, Kyriba’s positioning and cash forecasting centralises money from other sources, giving treasuries visibility across entities. Aiming not strictly at financial institutions, Kyriba’s offering makes sense for sprawling corporate entities with multiple operations.

Bank treasury teams looking for cloud-technology could turn to FIS’s treasury management system. The SaaS model does away with hardware infrastructure, IT call-outs, and has distinct benefits for remote workers. To protect against cyberattacks, FIS also offers private cloud hosting.

FIS Quantum Treasury cash management technology does number crunching, reconciliation and cash pooling. For bank treasury teams operating with multiple bank accounts, the platform can also manage bank accounts in tiers. Real time payment functionality or timed batch payments can also be made.

For treasurers looking for operational integration, Bloomberg’s partnership with treasury software provider FIS Integrity and Quantum makes sense. Bloomberg’s treasury management system takes orders from its trading platform and funnels instructions into the treasury management software. For major banking institutions already using the terminals, the ready-made integration means trades are executed on Bloomberg and captured directly onto the treasury system.

Still more ground to cover
To date, AI and new technologies are only starting to improve treasury operations. Wholesale automation across treasury desks is not yet a reality. “Using AI, treasurers can start to identify patterns across thousands or hundreds of thousands of transactions that they wouldn’t ordinarily have the resource or the time to do,” says Rohit Tulwar, CEO of think tank Fast Future Research to the Association of Corporate Treasurers.

“The tools will get better at spotting issues in real time and providing customised reports without moving data between systems. They will also enable far more automated production of reports and commentaries,” he says.

While AI and machine learning have huge potential to transform the industry, human decision-making will still prevail. The end decision is still with the treasurer to decide.  

Van der Poel says, “AI helps to execute certain processes and functions but they are designed and implemented by the control team and there are regular quality checks. People are still in control.”

Ultimately for banking teams, Kirk says, “Machine learning increases efficiency – and frees up time for treasurers to focus on more strategic tasks.”

Become a bobsguide member to access the following

1. Unrestricted access to bobsguide
2. Send a proposal request
3. Insights delivered daily to your inbox
4. Career development