The Rise of the Augmented CFO: Decision-Making is as Much an Art as a Science
CFOs and their teams have long been dedicated to supplying and analysing the data their companies need to make solid, fact-based decisions. However, finance departments have historically been constrained by basic forecasting techniques. Here Jean-Cyril Schütterlé, VP Product & Data science at Sidetrade, explains to Finance Monthly that CFO decision making, spending and innovating is […]
CFOs and their teams have long been dedicated to supplying and analysing the data their companies need to make solid, fact-based decisions. However, finance departments have historically been constrained by basic forecasting techniques. Here Jean-Cyril Schütterlé, VP Product & Data science at Sidetrade, explains to Finance Monthly that CFO decision making, spending and innovating is more of an art that we’re led to believe.
The underlying data collection process is often time consuming and error-prone, and the result frequently lacks depth, scope and quality. Not only is the underlying data unsatisfactory, but its processing is suboptimal. All of these approximate figures end up being copied from spreadsheet to spreadsheet and undergo many manual transformations.
This approach has many shortcomings:
- Regardless of the quality of the forecasting process, if the data is not detailed, sufficient, relevant and up-to-date, the result will be inadequate.
- Making the assumption that “all other things will remain equal” is an over-simplification. No lessons are learned from previous errors
Digitisation now gives access to more granular and diverse data about present conditions or past situations and their outcomes. Any data set that may help describe, explain, predict or even determine a company’s positioning can now be stored, updated and processed.
This 360° view provides an opportunity to discover correlations between the collected data and the figures tracked by finance executives in their modelling activity. But this trend line methodology is insufficient in itself to derive valuable knowledge from data diversity.
For the process of discovery to take place, this newly-found data trove needs to be mined with Machine Learning technology.
To put it simply, Machine Learning is the automated search for correlations or patterns within vast amounts of data. Once a statistically significant correlation is identified with a high degree of certainty, it may be applied to new data to predict an outcome.
Let’s take a simple example. Assume you are the CFO of a company selling goods to other businesses and you want to anticipate customer payment behaviour to prevent delays and accelerate total inbound cash flow.
The traditional approach would be to look at past transactions and payment experiences with every significant customer and infer a probable payment date for each.
But if you look closer at your data, you may find that your customer payment behaviours are not consistent across time, that your historical view is missing essential explanatory information about the customer’s behaviour that may or may not be specific to their relationship with your company. You end up shooting in the dark.
Wouldn’t your cash-in forecasts be much better if you had also correlated the actual time your customers took to pay you in the past, with detailed information about those transactions?
In theory, you cannot be sure that this model will perform well until you have run a Machine Learning algorithm on your own data, looking for predictive rules that relate each payment behaviour to the detailed information of the corresponding transaction or you have tested the predictive power of those rules on a set of examples.
In fact, the forecast is likely to be much more accurate than with the traditional methodology, provided that the data you fed the algorithm with were representative of your entire customer base.
That leads us to another question: can I find all this information about my past transactions while making sure they are representative?
Unfortunately, most of this information may not be readily available internally, either because you’ve never collected it or it is not flowing through your existing Order-to-Cash process. For instance, it is unlikely you know whether your customers pay their other suppliers late or not.
But SaaS platforms can capture most of this information for you and Machine Learning software will then be able to discover the predictive rules and apply them to your own invoices to forecast their likely payment dates.
But this is just a start. If inbound cash flows can be accurately deduced, so can other key metrics, such as revenue, provided the data is available. CFOs are the ultimate source of truth in an organisation. They manage skilled resources who translate facts into numbers and confer them credibility. They are therefore the best equipped to tap from as many diverse data sources as available, leveraging the power of Data Science to accurately forecast what comes next and thus gain marketing insight and competitive advantage for their company.
Thus, with their augmented capabilities, CFOs are now poised to be the digital pilots of today’s new data-driven organisations.