The Future of Corporate Treasury

16 Treasury and Trade Solutions Process improvement • Automate: RPA, APIs and machine learning (ML) will be used to automate repetitive, standardized treasury activities or processes such as bank reconciliation, cash application or exposure determination. • Connect: APIs will be used to facilitate connectivity between different solutions or platforms, with distributed ledger technology (DLT) being used in smart contracts (e.g., in the area of trade finance). • Learn: Use more advanced concepts of ML to analyze data patterns, learn and solve problems faster and more efficiently, which results in predictive solutions. • Assist: Make treasury technology more user-friendly and accessible, for example, via the use of a virtual desktop assistant comparable to Amazon’s Alexa, Microsoft’s Cortana, Apple’s Siri or Google Assistant. Data intelligence • Prevent: Analyze large sets of data to detect anomalies for fraud detection and payment outlier detection. For example, some vendors already use ML in their cash application solution to create rules based on detected payment patterns. • Predict: Analyze data patterns with the use of ML. For example, some vendors are developing smart algorithms to set up intelligent cash flow forecasting methodologies. • Prescribe: Use insight in data to create treasury and risk intelligence that suggests or even prescribes the next management actions to be taken. • Visualize: Envision large sets of data to gain insight into patterns, key developments and trends, to enable reporting on KPIs, etc. The future corporate treasury practice will see exponential technology and human interaction continuously working together, where technology will assist to automate and provide insight into data and humans will monitor, based on predefined limits and tolerance levels, and fine-tune where needed. Example: Operating the aircraft’s navigation and engine system Let’s make the above future vision more concrete using a simple example comparing treasury operations with operating an aircraft. The future airplane cockpit will use data intelligence algorithms that are continuously analyzing altitude, airplane speed, weather forecasts, etc. On the other hand, there will be a pilot (human or remotely controlled) who will take care of the procedures such as takeoff and landing, route changes, etc. The process of flying can be improved by using technology to: • Automate: Have the airplane fly on autopilot for long distance, mid-route periods. • Connect: Connect different airplanes, ground controls and satellites to always have a precise position. • Learn: Use statistics to understand the impact of crosswinds on the airplane. • Assist: Make controls easier to use and self-explanatory, improving pilot experience and reducing the chance of errors. • Prevent: Use weather data to fly around heavy storms. • Predict: Use wind data to predict periods of turbulence. • Prescribe: Provide suggestions to fly lower/higher/slower/faster or take alternative routes. • Visualize: Use flight data to build dashboards with route details, estimated fuel consumption, time of arrival, etc. This example shows that there will be a constant interaction between the technology with automated processes controlling the airplane and the pilot having the human interaction to validate the automated process and correct where needed.