‘Real-time treasury’ is an increasingly hyped phrase among certain banks and technology vendors, but how close to reality is this concept today? And how ‘real-time’ does treasury need to be? Four experts provide TMI with answers to these questions and more.
Over the past year, treasurers have been immersed in a series of interconnected global challenges, from supply chain issues to questions about energy pipelines – particularly in Europe – and a rising interest rate environment, coupled with inflation. As a result, timely data has been critical to the corporate response as treasurers look to insulate their firms from the risks involved.
To compound matters, companies that have seen explosive growth in recent years simply haven’t had to deal with this particular cocktail of challenges from a treasury perspective – at least in living memory. At the same time, there are traditional companies whose profiles and geographic reaches are entirely different now from what they were 10 years ago, so they are unsure how to respond, not having been in this exact situation before. There are also newcomers to treasury who have not been in the profession long enough to have witnessed anything like the scenarios currently playing out globally. All of this makes for a very uncertain operating environment.
Lorraine Donnelly, EMEA Sector Sales Head, Healthcare, Consumer & Wellness, Citi, outlines: “So many of these challenges that corporates are currently faced with have no playbook. That drives a hunger for real-time data to inform decisions, equipping treasury teams with the best possible information at the right time.”
The fundamentals
Three areas are essential for best practice real-time data management in corporate treasury. The data:
Must be accessible promptly
Has to be complete and therefore of good quality
Needs to be secure
The prospect of real-time data covering cash management, FX, payments, and risk management is becoming a reality through certain use cases. However, instant data across the scope of treasury operations is still some way off.
German Karaivanov, Vice President of Product, GTreasury, comments: “Most of the current sources of treasury data are near real-time. Think about bank balances, especially in a multinational corporation. The corporates deal with banks across the world, and some of these banks might not be operating in a way that is real-time for the corporate – they might be closed when the treasurer wants the data. Real-time data is about treasurers ensuring access to all the data being pulled, whenever they want it.”
Vital visibility and insights
Broadly speaking, corporate treasurers can leverage two primary real-time data sources: bank APIs and internal organisational technology platforms, such as ERP or customer relationship management (CRM) systems.
Kerstin Montiegel, Global Head Client Connectivity/Digital Client Access Channels, Deutsche Bank, notes that bank APIs represent the core enabler for real-time treasury. “Specifically, bank APIs provide up-to-the-minute account information data required for day-to-day treasury tasks such as cash positioning, but also to streamline payments,” she says. “On the company platform side, real-time ERP data, such as AR and AP ledger details, sales and purchase orders, and CRM data, such as sales leads, can help to improve the accuracy of cash flow forecasting. Combining this with real-time bank transaction and invoice data helps automate reconciliation workflows.”
There are various use cases for real-time data to be applied to treasury management. Data visibility in real-time is one of the most vital elements in areas such as instant payments, real-time cash visibility, real-time reporting, and real-time FX conversion and hedging. And the potential benefits are numerous.
“Global real-time cash visibility, improved reaction times, and reduced risks are a few of our clients’ use cases made possible through real-time,” comments Montiegel. “Real-time insights into cash positions help treasurers avoid unnecessary third-party financing, minimise interest expenses, leverage the available cash for investments, and increase their company’s net income. Additionally, real-time FX conversion and hedging maximise the benefits of real-time payments for corporates working in multiple currencies.”
Loic Leonard, Head of Product (Wallstreet Suite), ION Treasury, agrees. “Bank account visibility enabled by real-time data will empower treasurers to manage their cash cushion more aggressively,” he says. “Whether they have €100,000 or several million euros sitting in their account, as the treasurer gains a far clearer intraday picture of what’s going on, they can start to reduce this buffer, effectively freeing up additional cash. They can grow the business without needing more working capital or additional funding, and of course, if they fund less, there is less interest and fees to pay.”
Data intelligence offers another ‘instant’ use case. For example, cash flow forecasting enhanced with real-time data helps treasurers to maximise liquidity allocation and risk management. Montiegel notes: “Real-time data also leads to better working capital management and can help treasurers free up trapped cash. Also, bank APIs enable the automatic pre-validation of beneficiary account details before payment execution, meaning that treasurers can save time and avoid investigations into failed payments and fraudulent cases.”
On the risk management side, slashing the time data takes to travel between different systems is essential, particularly when managing FX. Leonard explains: “Corporates are increasingly using trade venues to execute transactions such as FX hedging, for example. The less lag there is between when a treasurer trades on the venue, such as Bloomberg or 360T, and when they see that reflected on the TMS is critical. Treasurers see their risk netting out and reflecting their hedges.”
This is particularly helpful when treasurers are using several trading venues because the venues will offer different prices or liquidity on different currency pairs, for example.
“The TMS should aggregate all these different data flows with API integration, presenting the user with the complete picture,” adds Leonard. “Using real-time data in this way is a very positive step because it enables treasurers to manage their intraday cash position based on their risk.”
In addition to the typical bank notifications, another use case for real-time data is in data processing, where corporates can establish real-time processes that affect treasury management, as Montiegel explains. “Electronic bank account management [eBAM] via API is a concrete example of a high level of reliable automation within a treasury process,” she says. “This can be enhanced if it is combined with robotics. More reliable information and fault-tolerant processes allow for better scenario automation. Real-time data processing is an enabler for smart logistics, smart cities, and smart supply chains, all of which are a reality today.”
Inspiring interconnectivity
Of course, accessing real-time data for treasury works only if internal stakeholders and external partners are on the same page. Crucially, the treasurer can set the tone of how these various inter-relationships work. This starts with the treasury team deciding what type of real-time data infrastructure they want.
“Treasurers have a budget, IT resources, and a specific goal of the data they want to collect,” Karaivanov says. “To create the best infrastructure to collect this data, there are three options. They can either build internally, which will require IT resources and funding; buy using a consultant to help them create a solution; or partner with a TMS provider. All these internal factors come into play when deciding what’s most cost-effective and the fastest way to go live, to access the data, and start making decisions based on it.”
Whereas legacy bank connectivity channels, such as host-to-host connections, did not enable real-time data exchange, treasury software vendors and banks have started to build partnerships and applications that leverage the emerging bank API infrastructure.
Montiegel advises: “Treasurers should reach out to their banks and TMS providers on their API journey to see how they can help transform traditional treasury processes into the era of real-time. When treasurers are looking for a concrete collaboration with different stakeholders internally, they should focus on how treasury obtains the data and how their colleagues make it available. Often this will involve upgrading a legacy system or processes to a certain level to leverage real-time data availability.”
Externally, most banks have been working on IT knowledge and support to establish API connections for treasurers. Many have focused on offering treasury-critical real-time solutions into a single API management interface. However, the service level across different areas of treasury can vary from bank to bank.
“For around the past five years, we have seen this desire from the banks to differentiate themselves from their competitors in terms of speed of service,” notes Leonard. “One bank may offer instant payments, while another will offer real-time balances, and so on. While the technology they are all using is relatively similar, it’s not standardised. Corporates that want to talk to their five core banking partners need to establish APIs and have a good relationship on the technology side with each of them. This could be a challenge.”
Banks aren’t developing their API portfolios on an island, however, with some notable collaborations with fintechs in the past year, including Goldman Sachs’ partnership with Modern Treasury.[1] The goal is that fintechs can leverage their bank partnerships to standardise infrastructure connections and streamline data access across multi-bank offerings. However, standardisation in the world of treasury API partners is somewhat lacking.
“In terms of external parties such as trading venues and market data providers, each provider will tend to have their specific API,” states Leonard. “Even though some protocols are in place, each API has a different flavour. It’s like SWIFT messages – they’re supposed to be normalised, but in reality, every bank is doing something slightly different. So, corporates need a powerful transformation layer in their TMS.”
Getting the tech right
The dream of real-time data for treasury will become a reality only if the technology in the treasury department and the partners it requires the data from is fit for purpose. TMS vendors and other service providers linked to banking services have tried to establish themselves over the years. Many of the building blocks of treasury services offered by the TMS providers today leverage real-time data. This has been enabled by three critical drivers: progressive regulation such as PSD2, advanced technology including AI and ML, and the popularity of instant payments.
“It is important to understand that PSD2 APIs are rather limited in scope and functionality,” comments Montiegel. “Treasurers should ask their banks and TMS providers about capabilities that go beyond that to assess their API maturity. Today, increasing numbers of treasurers are already taking advantage of more advanced banks and TMS providers that provide services such as real-time intraday sweeping, FX conversion, hedging, and cash visibility.”
In the past, early movers were the primary drivers of API adoption. What banks are experiencing now is significant demand by large, multinational corporates. “At the same time, TMS providers are approaching us to provide their clients with plug-and-play API connectivity,” continues Montiegel. “These are all indicators that it will be just a matter of time for real-time treasury to cross over into the mainstream.”
The holy grail for treasurers is an integrated multi-bank platform where individual solutions can be easily consumed and combined with data from other service providers and banks. Individual solutions must be able to support ease of implementation through strong integration and data extraction capabilities. They must also support solution portability across multiple banks in a bank-agnostic environment and address customisation issues that make adopting solutions difficult, such as extracting data from diverse IT landscapes.
Once the plumbing is in place, technology supports the analysis that treasurers need in a real-time environment. This is where the likes of AI and ML can start to pay dividends, given enough of the correct data.
“Treasury can apply AI to process real-time data and visualise data on the timing of payment runs, historical patterns, to analyse customer payment behaviour, and create unprecedently fast and accurate forecasting,” outlines Montiegel. “It can also enable a real-time view of liquidity held across bank accounts combined with short-, medium-, and long-term cash predictions across currencies and legal entities. Analytics tools and AI solutions can generate data-driven insights at a group and legal entity level. A better knowledge of future cash flows also enables treasurers to reduce their cash buffer and benefit from more efficient funding structures.”
AI models can also support ‘anything as a service’ (XaaS) workflows that combine different tools to automate a process end-to-end. Montiegel brings this concept to life in an example where a company uses an IHB solution combined with a cash pool structure. Here, pooled funds can be automatically invested (based on company policies and instructions) into bank deposits in multiple banks and some MMFs. The bank accounts used for payments are filled considering the different cut-off times of the regions to avoid overdrafts and excess funding.
“Real-time technology and AI can also be combined to execute hedges based on user-defined rules,” she adds. “For example, Maestro, Deutsche Bank’s hedging solution, can respond to exchange-rate fluctuations in real-time. However, this also depends on how quickly a treasurer’s systems can feed information into the platform.”
A fresh perception
While the path to mastering real-time data for treasurers has been charted, it is not free from obstacles. Indeed, one of the critical challenges lies in managing expectations, particularly around technology.
“If treasurers think that they can feed AI and ML any data and the technology will provide an accurate prediction, they will be disappointed,” warns Karaivanov. “Treasurers must have a lot of data to feed into an ML model to get a good prediction out of it. Infrequent data from just a few days in a month or a year won’t support an AI or ML predictive model.”
Quantity of data is also an issue when considering the data flowing into the treasury before any analytics is applied to it. Not all data will be essential to treasury operations.
“Treasurers moving into the real-time data world is similar to the average person moving from traditional TV and radio news to the internet and being suddenly inundated with millions of pieces of information,” Leonard says. “Treasurers need to know how to consume data at this scale, or they will be completely overwhelmed. Treasury technology must be ready and designed for real-time data.”
Another area where treasurers need to look under the hood is with their external partners’ technology – does it do what they’re being told it does and is it right for their needs? Overselling of APIs can be a culprit in this regard.
“A lot of the time, treasury partners such as banks or ERP vendors will say they have an API, but it will turn out that it is not mature enough to be utilised properly,” posits Karaivanov. “It can have some volume limitations as the amount of data flowing through the API may have been underestimated. That impedes some of the integration work being done to enable that data from partners to be pulled into the treasurer’s dashboard.”
These types of kinks must be ironed out if the concept of the real-time treasurer is to become a reality.
Evolution rather than revolution
For treasurers exploring how best to apply real-time data to their role, there are many considerations. For Leonard, a treasurer’s approach to real-time data should stem from their needs.
“They should compare their treasury operations to those of their peers and see where they could be more efficient,” he advises. “Usually, this will come down to needing more STP and automation, enabling treasury teams to operate at the same level with fewer people or do more with the same amount of staff. Defining the areas for improvement is vital because not everyone needs real-time for everything. When this has been identified, the treasurer must seek the right TMS to accompany them on this journey. They could already have it or they may need to look for a more powerful system with real-time capabilities.”
Another consideration for treasurers is understanding how real-time data may disrupt various treasury processes. A vivid example of this is how treasurers manage real-time payments. Karaivanov reflects: “Payments are not simply about sending out the payment. It’s assessing any potential fraud risk or any secondary approval that needs to happen from internal colleagues before the payment is sent out. Real-time data for payments may not necessarily be viewed as a good thing by corporates because it can increase their fraud exposure. Having the best processes in place for the real-time environment is essential.”
The exciting news, though, is that examples of treasurers adopting real-time data flows, while rare, are increasing. These activities can guide others on their journey.
“Think of real-time treasury data as an evolution rather than a revolution,” Montiegel concludes. “Treasurers can achieve a lot with the existing API infrastructure today. An increasing number of large multinational corporations recently started to take the first steps, which means the next wave can build on the experiences made by others. We are about to see the shift where API connectivity and real-time treasury become the new normal.”
