Bank of America today announced the launch of CashPro Forecasting, a tool that uses artificial intelligence (AI) and machine learning (ML) technology to more accurately predict future cash positions across clients’ accounts at Bank of America and other financial institutions.

The solution was developed in collaboration with a fintech that specializes in applying machine learning to cash forecasting to help solve a top pain point for companies: how to measure future cash needs without a significant manual effort or a costly technology investment.

Ken Ullmann, co-head of Global Commercial Banking, Global Transaction Services at Bank of America, comments:

“Many companies today rely on manual, repetitive work to forecast their cash needs, leaving little time to analyze the data for making strategic decisions, which is the actual objective of the forecasting exercise. With CashPro Forecasting, companies can automate their forecasting process while improving the accuracy of their predictions, all without making any IT investment.”

CashPro Forecasting is available to clients already using CashPro®, the platform that the bank’s corporate and commercial clients use to manage their everyday treasury and borrowing needs.

“Unlike competitor solutions, CashPro Forecasting requires no onboarding since it’s fully integrated into the platform, allowing it to automatically plug into the client’s historical cash flow information,” said Jay Davenport, head of Product Structuring Sales in Global Transaction Services at Bank of America. “A client can activate the service themselves from their CashPro settings, and begin using it immediately.”

CashPro Forecasting uses a variety of ML models to learn from a client’s historical cash flows, automatically selecting the most accurate one for each account and using it to forecast future cash positions. At the end of each day, the models are retrained based on the new cash flows from that day, allowing them to continuously learn and adjust for changes in seasonality and other operational impacts specific to each company.

The key features of CashPro Forecasting include:

  • Improved accuracy – Machine learning technology automatically learns from historical cash flows;
  • Scenario analysis – Users can create their own forecast scenarios by taking trailing averages or assigning growth rates;
  • Multibank reporting – The service forecasts cash flows across Bank of America and bank accounts held at other third party banks;
  • Transaction mapping – Users can create simple rules to automatically classify and forecast individual transactions into categories such as Collections or Payroll;
  • Simple setup – Zero implementation. Companies can enable the service directly within CashPro and begin using it immediately;
  • Customizable – Companies can forecast by account, currency, transaction type, or choose their own cash flow categories. These can be done on a daily, weekly, or monthly basis, up to a year in the future.