The accounting world will never be the same again. The advent of big data, data analytics, cognitive computing and the internet of things (IoT) is seeing chief financial officers (CFOs) and financial controllers move from managing to actually running the business. In the future, finance will no longer be a support function, it will be a core decision-making department centrally involved in business strategy.
Finance departments are in possession of vast amounts of data relating to the business but their traditional role has been to crunch the numbers and do little else. The new wave of data analytics technologies allied to advanced cognitive computing is bringing about quite radical changes to that traditional model.
In its "How CFOs can own analytics report", Deloitte points to a future where CFOs will not only be involved in planning, budgeting, and forecasting but also in day-to-day, or even minute-to-minute, operational decision-making in areas such as production and sales and marketing.
This is because the CFO now has access to information from across the business, including what customers are buying and when, what supplies are being purchased for what departments and for how much, and precisely how much is being spent on marketing budgets, rents, travel, and everything else in the organisation.
“Armed with analytics, CFOs can exercise more centralised control of operational business decision-making, answering questions such as, what price point should be used for this customer on this day, or what inventory products should be pulled forward or out of the supply chain?”, the report states.
It goes on to point out that advanced analytics can allow companies to hedge against volatility and to respond faster, and with greater insight, to changes in the marketplace, to such a degree that the predictive power that analytics offers in pricing, the supply chain, and other areas can have an increased impact on how business is done.
Adding IoT to the equation places the CFO in an even more powerful position. The real-time information on key metrics from production processes and distribution networks delivered by this technology offers the potential for enormous cost savings and efficiency gains.
KPMG director Kieran O'Brien points out that people have been saying for many years that CFOs were going to move into this more strategic role but that the difference now is that they have the tools to make it happen. "They have been collecting the data for many years but now with big data and predictive data they can get behind the numbers and look at the drivers of profitability. A lot of strategic decisions are driven by understanding performance. Using big data a CFO can see what customers are the most profitable and what products are most profitable. They are not guessing anymore and they can decide whether certain products need to be dropped or not, for example. Can the firm do with two types of sliced pan instead of three?"
Real value
According to the American Institute of Chartered Public Accountants (IACPA), the real value lies in predictive and prescriptive analytics. Many people consider data analytics to be limited to descriptive analysis, the explanation of what has happened. Predictive analytics covers what will happen while prescriptive analytics looks at what should be done in response. Predictive analytics takes data from various sources within the company, integrates it and uses it to predict future outcomes based on historical and other data.
O’Brien gives the example of aircraft leasing. “We know that passenger numbers are going up but that’s not enough,” he says. “We need to know why and where they are going up and be able to predict what kind of aircraft – narrow or wide-bodied – will be most successful and what the financial impact of that will be.”
Mazars audit director Emer O'Riordan believes that CFOs are now in a position to move from mere diagnostics on to more valuable predictive analysis but there is work to be done. "CFOs already rely on analytical processes to drill down into queries where results are somehow different from expectations," she says. "Data management and a good technology infrastructure that gives timely information that is easily visualised and understood by the user may already be a challenge for CFOs with legacy systems, or groups with incompatible systems and key performance indicators (KPIs) that don't have a common data platform to work with."
Those with more up -to-date systems are in a different position, however. “The CFO with a good platform of data who has worked on making data collection and reporting more efficient and automated, has a golden opportunity to use data analytics to move from a diagnostic to predictive model to help get insights, make better decisions and really impact the bottom line,” O’Riordan adds. “It is also interesting that where predictive modelling has an impact on behaviour in an organisation, where it starts to learn the drivers of something like non-compliance that the business process adapts in a preventative way reducing time spent by finance to drill down into problems. This feedback loop has a very powerful impact on overall efficiency.”
The potential benefits span the entire business, she believes. “Anywhere the CFO can identify indicators underlying a problem, they can use predictive analytics to model these as potential or common predictors of where this challenge will arise again. Working capital management, predicting customer disputes and maximising discounts all impact on the bottom line. Judgmental areas such as provisioning can use predictive models to give more robust support and accurate estimates. Customers and margins can be predicted giving insights to make better customer segmentation decisions.”
But these systems must be kept constantly updated and users need to understand that they are not infallible. “Any predictive model cannot stand still, but must learn from its mistakes and the changing environment,” O'Riordan points out. “The CFO should have a handle on assumptions used, regularly review their appropriateness, and ultimately understand the limitations or conditions where the model may not be accurate.”
Advanced data analytics: Introducing the robo-auditor
The auditing process for companies of all sizes, but for SMEs in particular, could be made a lot simpler as a result of advanced data analytics. Experts already foresee a time where an SME’s annual accounts can be produced at the touch of a button. The system will be able to gather all of the relevant data input during the year and assemble it into accounts which comply with company law and accounting standards ready to be presented to auditors.
Of course, larger and more complex organisations will not find it quite so easy but will still profit from the overall efficiency gains.
The auditing process will change as well with the new technology doing much of the heavy lifting. “Modelled behaviour can reduce a large population of transactions to be reviewed into a smaller pool of higher risk transactions where energy can be focussed,” says Mazars audit director Emer O’Riordan. “This is already impacting on audit methodologies, where data analytics are being used to identify high risk transactions and populations by using modelled criteria rather than random sampling.”
We may not quite be at the stage of a fully automated audit but we could be on the way.