Advancements in technology are driving innovative solutions that enable finance and downstream users to better leverage financial data. This includes technology such as artificial intelligence (AI), machine learning (ML), deep learning (DL) and blockchain. Some of the promising advancements include:
- Leveraging AI, machine learning and deep learning to better predict future results will introduce more internal and external data into the forecasting process to predict future sales, profitability and cash flow based on micro and macro-economic factors. Forecasts leverage learning models to compare predictions with actual results and identify factors that have a high correlation to the final numbers. These factors can be incorporated into the models to improve the ability to predict future events. Additionally, machine learning can be leveraged for cluster analysis to determine common behaviors in accounts receivables, vendor performance and customer purchase behavior.
- Leveraging blockchain technology to better trace activity like transactions, payments and supply chain flow enables quick traceability and provides insights into specific events as they occur.
- Automatic alerting platforms scrub financial and non-financial data to alert users to specific activity that may impact their business unit, department or overall organization. These AI interactions enable the users to understand, in real-time, transactions that fall outside of preset rules and thresholds customized to the organization.
- Fraud detection models leverage audit automation technology to score and identify potential fraudulent transactions as they occur. This uses much of the same technology currently deployed in the banking industry for credit and debit card fraud. Examples include developing a customer risk factor to determine payment patterns that indicate a customer’s likelihood of default for use in pricing and contracting and calculating an employee risk factor to identify “out of compliance” spend and irregular purchase activity.
The CFO of the Future will have access to many tools that provide the value-added analytics required by the business. The question is, will they have the skill sets to use them? The position of data scientist will become a critical component when it comes delivering quality CFO analytics.
The data is already in place to feed predictive models, but having the data is only the beginning. To truly transform the CFO function into a modern provider of predictive and prescriptive analytics, several business questions must be answered, including:
- What tools are required to develop analytics that will better support the business?
- What are the business outcomes that need to be created by new analytics?
- What is the business value which will be generated?
- Are the right resources and skill sets available?
- Is a data scientist required?
- Are the required data sets available to support this?
- Is the data clean?
Development of new analytics for the sake of driving new insights is a failed strategy. The effort must be supported by the business value created. Once the value is established, the effort to develop new models that are trusted, repeatable and will drive a return on investment can begin.
Analytics that will provide the users with business value added insights fall into two key categories:
Predictive analytics leverage internal and external results and other factors to predict what will happen in the future. It is useful in understanding what will happen based on specific changes in the core variables and will result in better forecasts of revenue and the identification of cost saving opportunities. Predictive analytics also results in a better understanding of the key factors that have a high correlation to success. Better decisions results in improved sales and bottom line profitability.
Prescriptive analytics enable the users to take actions that can improve business decisions. Prescriptive analytics tools like path analysis, machine learning, artificial intelligence and data science driven models detect fraud and compliance issues, provide interactive forecasting models and present interactive AI based on business rule driven decisioning options. Analytics can predict the business value impact each decision opportunity presents.
So – what does the CFO of the Future look like?
The CFO of the Future uses more predictive and prescriptive modeling to demonstrate value added business driven insights and provide the best return on investments. New analytical techniques better predict future results and are continuously updated and adjust based on past predictions using machine learning techniques. The models highlight various paths that occur and how to prevent taking paths that drive negative results. The great news is that this can be based on the traditional detailed datasets that have years of history.
Finance analytical users are often laggards when it comes to new or emerging technology, preferring the wait and see approach. This may explain why blockchain has yet to emerge as an option in many companies. Blockchain may still become accepted, but the typical finance viewpoint is to wait for emerging solutions to mature.
The CFO of the Future will be more strategic and analytic in nature. Building a core, detailed-driven foundational layer will open the doors to leveraging new analytical tools that can provide innovative, yet repeatable solutions for the business. We are just getting started, but the future of AI, ML and other more sophisticated analytics leads to more consumers of financial data and enables better business decisions. Finance will be the “heartbeat” of reporting and analytics for the company.
In the next CFO Analytics series, we will define machine learning, show how it can be leveraged in finance and reveal the benefits it can bring to end users.
