Robotic process automation (RPA) and machine learning (ML) are accelerating the pace of change in corporate treasury, leading to better decision-making and enabling treasury practitioners to focus on strategic, value-added activities.

RPA uses a software robot that mimics human actions to automate repetitive, tedious and high-volume tasks and processes faster and with more accuracy and efficiency. Unlike RPA that is associated with “doing”, ML is concerned with “learning”.

ML is a subset or branch of artificial intelligence (AI), in which a machine learns how to identify patterns in data. ML is used to analyse data to identify trends, find patterns or make predictions via the use of algorithms. In essence, ML gives machines or computers the ability to learn and make data-driven decisions without explicitly being programmed.

Automation is playing an increasingly important role in strengthening an organization’s competitive advantage, and emerging technology is becoming critical to improving operational efficiency.

The Association for Financial Professionals (AFP) Treasury in Practice Guide, Identifying Value for Treasury: Automation, Machine Learning & Artificial Intelligence, underwritten by Kyriba, outlines the benefits of RPA and ML and how each is being used to solve problems faced by individual treasury departments. In addition, the guide shows how to build a strong business case when seeking to adopt RPA or ML.

Here are the important topics and content presented in the AFP guide that will help treasurers prepare for treasury technology shifts, challenges and opportunities, and take advantage of the advances in these two disruptive technologies.

Potential benefits of RPA and ML

Although the use cases vary, RPA and ML offer similar potential benefits, including 1) improved and consistent levels of accuracy given that they are less prone to human error and repetition fatigue; 2) significantly reduced processing and decisioning time as bots and machines carry out actions quicker and can perform routine or mundane tasks in a fraction of the time it takes a person to complete; 3) results and calculations available, globally, when needed since ML and RPA can operate at any time; 4) better time management and productivity because automation eliminates repetitive and manually intensive tasks, letting treasury employees allocate more of their time on higher-value problem-solving work; 5) and improved morale due to reduced stress of calculating positions under time pressure and decreased risk of error, enabling treasury teams to devote more time to value-added and rewarding activities.

How RPA and ML are being used to solve treasury problems

This section in the AFP guide outlines three specific ways RPA and ML are being used to solve problems faced by individual treasury departments.

1. Automating time-consuming tasks

An insurance company’s treasury department would spend hundreds of hours a year processing internal customer requests for check images. These requests were time-consuming to complete and added no value within finance.

The treasury team developed an internal RPA process to automate the repetitive and time-consuming task. While the team had to spend some time training their internal customers on the new process, the system is now operational and running three times a day.

Using the bot has improved the response time for treasury’s internal customers. In addition, the customer experience and the service level agreement (SLA) have improved, as treasury can now respond to a customer’s request in a matter of hours, rather than weeks.

Adoption of RPA has also helped release time back to treasury to focus on activities that add value.

2. Improving cash flow forecasts

RPA is a useful tool to improve cash flow forecasts and can be leveraged to process data from a number of different company systems – data sourced from banks, treasury management and enterprise resource planning (ERP) systems and from other company departments, including payment teams.

Séverine Le Blévennec, former senior director of EMEA Treasury at Honeywell, led the development of an RPA process to improve the accuracy and timeliness of their in-house bank’s cash flow forecasts.

Le Blévennec reviewed all existing processes and created seven workflows for seven different activities, from maturing time deposits to intraday payments and collections. She also emphasized the importance of testing.

Tests were performed with a mindset that “things could go wrong.” Then, before going live, the treasury team used the new spreadsheet alongside the old process to ensure everything worked as expected.

Since going live, Honeywell has witnessed some significant benefits. The new system saves time and is more accurate and efficient. The old manual system could only include data from about 40 Honeywell bank accounts, but today data from over 160 accounts is included in the cash flow forecast, and any new accounts can be added easily.

Additionally, enhanced visibility has improved counterparty risk management, reduced levels of uninvested cash and, as a result, increased investment returns. Honeywell’s treasury team is now less stressed and more engaged in their digitization journey.

3. Better receivables management

A technology company’s accounts receivable (AR) team was witnessing a steady growth in sales, resulting in increased receivables. They were not pleased because they were so busy processing payments that they didn’t have the time to convince customers to transition to more efficient, less costly electronic payment formats.

Increased sales and over 2,500 monthly checks to process, all needing some form of human intervention, led to delays in receivables processing and decreased confidence in the accuracy of AR data in the ERP.

The treasury team recognized the need for change. The team did research and spoke to their banking partners on how to implement a more modern lockbox, utilizing the optical character recognition (OCR) codes. The discussions ended with the team realizing that ML could be used to automate some of the processes.

The result of designing and implementing the ML receivables solution was dramatic. By replacing the manual gathering, consolidation and formatting that was required every morning, the AI solution enables the company to speed up the payment process. Given that the information is consolidated and the payments are standardized, reconciliation is simpler, and most payments are processed within two days.

Three months later, the team has a faster response to queries, and information is tracked immediately, removing the need for time-consuming searches for data. The receivables solution is also being used to match payment to remittance, and also to provide better remittance information to customers, further improving matching. Moreover, ML is now freeing the treasury team to spend a greater portion of their time on high-value work.

Building the business case to adopt RPA and ML in treasury: five key considerations

The AFP guide recommends that as with any technology project, it is critical to build a strong business case when seeking to adopt RPA or ML.

To help make a compelling business case, there are several key considerations. First, understand the technology to maximize the potential benefits. This implies identifying the pain points or the nature of the treasury problem to be solved, understanding these technologies in some detail, ascertaining how they will solve the particular problem, and matching the technology to the task to help maximize benefits of RPA or ML adoption. For example, if you want to automate a process, RPA is likely to be the more suitable solution. If you want to analyse data, ML will be more appropriate.  

Second, optimize processes before automating. This means mapping and reviewing existing processes to determine if they can be made more efficient.

Third, communicate and educate stakeholders on the proposed solution. Communication is integral to the success of any project. This involves engaging stakeholders early in the planning process to seek their support and secure resources needed for the technology project. Getting buy-in from a treasury leader or project sponsor and approval from all required stakeholders will give credibility to the RPA or ML project.

Fourth, identify potential returns. This means determining the measurable costs and benefits of replacing mundane processing and time-consuming data collection and analysis with RPA or ML adoption, including quantifying the softer benefits in the form of released time and reduced risk of error and fraud.

Fifth, establish and monitor success metrics. This indicates the importance of recognizing some clear targets as measurements of the success of the project, such as how much time an RPA project will save. It may also help to illustrate consequential benefits, such as improved investment returns due to more accurate cash forecasts.

Finally, scale the solution. As technology evolves, there will always be more ways it can be embraced to improve treasury operations and look for the next step. Measuring the outcome of one disruptive technology project will help augment support and momentum for future development and assist treasury practitioners in introducing more technology transformation initiatives.

Conclusion

RPA, ML, AI and other breakthrough technologies will play a more prominent role in the efficient management of corporate treasury departments in the coming years.

Visionary treasurers will leverage the best of emerging technologies to deliver smarter treasury insights, achieve operational and liquidity effectiveness, mitigate risks and reinforce treasury as a strategic partner to the wider business.