The Third Wave of Financial Automation – CPM + Artificial Intelligence
Prophix Oct 10, 2019, 7:23:00 AMIn my last blog “All AI Paths Lead to the Cloud,” I talked about how the FP&A challenges facing finance leaders are not going away. These issues are only compounding due to an overabundance of data and the rapid evolution of hyper-connected mobile employees, driving businesses towards the availability, scalability, and affordability that comes with putting their financial applications to the cloud. The era of simply throwing more people and resources at the challenges simply does not economically scale. Why? Well let’s think about the problem:
- We are gathering gobs of information faster than your analysts can effectively process
- Analysts are spending more time searching for data than they are analyzing it
- This leaves finance management without the insights they need to confront their business risk head-on
- Incomplete analysis leaves room for loss and risk
Analyzing Data with AILet’s look at the challenge of an ever-increasing amount of data and the lack of resources to effectively analyze it. Specific to FP&A, we will use the month-end close process as an example. If you are a mid- or large-size company doing hundreds of thousands or even millions of GL transactions per month, it is theoretically impossible for your finance team to comb through each transaction looking for all anomalies or possible fraud. So, when things don’t balance, your month-end close is typically delayed until you can root out the serious issues. These delays are costly and introduce a window of risk. But, what if you could analyze 100% of your financial data and automatically see your suspect transactions in real-time? Before the close process? Now you can by leveraging the intelligence and power of Machine Learning. Definition: Machine learning (ML) is the scientific study of algorithms and statistical models that computer systems use to perform a specific task without using explicit instructions, relying on patterns and inference instead. Having your suspect transactions instantly identified for you removes the need for human analysis while enabling human action to resolve anomalies. Machine Learning also continues to learn your business, making it more efficient and accurate each time you use it. Some questions you might ask yourself are:
- How much time would that shave off your month end?
- How many more questionable transactions would be revealed versus your manual spot checks?
- How much more loss or potential fraud would this eliminate? - Humans are currently only catching about 5% of loss as per the 2017 Total Gross World Product Loss.