For many CFOs, the time is now to embrace AI for cash forecasting
The chief financial officer (CFO) has never been under as much pressure to deliver more accurate cash forecasts – the anticipated revenue, spending, and liquidity data that acts as the rudder for all corporate decision-making. More precise foresight is essential not only to driving profitability under normal business conditions, but has now become even more crucial as companies try to navigate the continuing wake created by COVID-19.
The challenges are clear, and there has never been a more opportune time to incorporate artificial intelligence (AI) and machine learning (ML) strategies into a CFO’s playbook.
AI and ML capabilities have matured to a point where there are now clear and proven use cases for CFOs to introduce much-needed automation, efficiency, and improved accuracy into forecasting processes. By leveraging AI and ML, CFOs and their teams can effectively automate modeling across multiple scenarios and utilise data sets magnitudes larger than would otherwise be possible.
Freeing their finance and treasury teams from manual forecasting work also allows the CFO to focus the treasury team’s efforts on initiatives that deliver more value for the organisation. In doing so, AI transformations result in expediting forecasts and achieving analysis with a depth and precision that go far beyond the capabilities associated with manual processes.
Still, McKinsey finds that only 33% of organisations are actually leveraging AI capabilities effectively. Considering the complexity of navigating the current business environment and ongoing economic uncertainty, CFOs should rightly embrace AI-based forecasting strategies as not just advantageous now – but also all-but-required going forward.
Where manual forecasting falls short
When only manual capabilities are available, financial forecasts are often limited in scope and may struggle to line up with how reality ends up playing out. Too often, CFOs and other corporate leaders are dissatisfied with the results – a waste of time at best, a cause of ill-fated business decisions at worst.
More specifically, forecasts produced manually are often poorly integrated with sales and operations forecasting. This leads to findings that are out-of-date by the time they’re prepared. Manual forecasts often use rudimentary models that simply multiply current results by subjective growth factors, rather than incorporating business data and drivers for more effective projections. Manual processes that involve spreadsheets also make it difficult to keep data sources consistent across business units (while being significantly time and resource intensive).
Given these and other challenges of producing and updating forecasts manually, CFOs and their teams are less able to investigate multiple scenarios and can struggle to deliver timely and accurate insights when it matters most.
Incorporating AI effectively eliminates the challenges of manually compiling forecast data, as well as the limitations of spreadsheet-based processes. Whereas manual data input and imports can produce errors and inconsistencies, AI technologies can store and compute data with far, far greater accuracy and speed. This is especially true when AI tools are incorporated natively within treasury and financial management systems, automating connectivity and data availability across an organisation. Ideally, the systems will offer the simplicity and consistency of utilising a single data set across all modules to take full advantage of the AI capabilities.
AI automation similarly enables larger data sets to be collected and put to use. At the same time, AI algorithms learn and iteratively improve the accuracy of forecasts far more quickly than can be done by manual means. As AI tools take over these mundane and often error-prone analysis tasks, CFOs and their teams are able to redirect their efforts and work more efficiently, focusing on higher level analysis of operational flows, economic drivers, and business strategies.
Finding the right AI strategy
CFOs committed to an AI-fueled strategy for financial forecasts must decide whether their existing use cases suggest any realistic potential for building out in-house AI implementations, or if they should instead enlist an external solution. Collaboration with technology leaders across departments to assess internal AI experience and talent is essential to this decision.
In cases, where other teams are already engaged with AI tools or API connections into such tools, it may make sense to expand those projects into financial forecasts. Some organisations take a hybrid approach as well, leveraging both internal capabilities and external solutions. Cost, dependency on outside departments, and time-to-value are all influenced by the chosen route.
But to be clear, CFOs that plan to use AI for forecasting alone and that cannot piggyback off other internal projects should absolutely go shopping. Internal costs for developing capable AI tools commonly reach into the millions of dollars – and require a team of data scientists and months of algorithm training to utilise effectively.
Tools come in two varieties: those specifically focused on AI (often newer fintech start-ups), and those with existing cloud-based software recently enhanced with AI capabilities. Specialised AI vendors offer advantages in the areas of time and cost savings. However, the risks of working with start-ups must be considered, as well as a tool’s potential longevity and the long-term support it will be able to offer.
Alternatively, CFOs able to work with their existing systems and providers to add AI functionality can often get up and running faster and take advantage of historical data, training, security, and compliance utility already in place. A survey by Deloitte found that 60% of C-suite decision-makers opted for this strategy leading Deloitte to name it “the easiest path: using enterprise software with AI ‘baked-in’”.
CFOs that champion their internal AI transformations to power forecasting processes have the opportunity to introduce new tiers of speed and agility, while enhancing their treasury and finance teams’ value to the organisation.
The right AI strategy can maximise the advantages of automation and offer a near seamless implementation.
By providing faster forecasting analysis of larger data volumes within a framework of unified business systems, AI is more than ready to enable that precise accuracy and depth of foresight that organizations now know is a necessity.