A large, multinational technology company wanted to transform its corporate finance functions, starting from revenue forecasting to compliance and auditing functions. The processes include the use of Excel sheets for exchanging data, which was mostly manual. The company was dealing with a large volume of transactions, which required sampling for validation. There was a need to score, flag, and predict high-risk deals through measurable preventive means.  

Neal Analytics created Machine Learning based ensemble models to extract the “seasonal” patterns and forecast. With this solution, the company could apply external input (exogenous variables) to adapt the forecasts for new information (e.g., search engine meta-data). Pattern recognition for fraud or abnormalities can be applied to the entire data lake constructed from finance transactions. It helped them optimize decision making in the presence of uncertainty and error, thus forecasting risks.    

Using machine learning capabilities, the company’s financial forecasting became faster, better, cheaper, and more transparent. Automation led to reduced lead-time required for the process to two days before cut-off. It has also resulted in increased accuracy, which is 2-3 times than the manual method. The full-time equivalent (FTE) coverage was dropped to 25% of the original value. By automating processes, the forecasts were standardized and documented for accountability.