You Are Not Alone: Common Problems with Cash Flow Forecasting
We all know why cash flow forecasting is important, but it’s still worth reviewing why we all invest so much time and effort into it. Accurate cash flow forecasting is a critical element of finance’s role. It helps companies ensure they have enough cash to meet their obligations and can comply with any lenders’ covenants. It also minimizes the risk of having to borrow unnecessarily or at short notice, and helps companies plan short-term investments. Crucially, accurate cash forecasting helps finance play a strategic role in their businesses.
We at Tesorio spend a lot of time talking to CFOs, and it’s clear that accurate and timely cash forecasting is more important than ever in these uncertain times. More than a few of our customers have told us they are now updating and reporting on forecasts weekly. Our conversations have also made it clear to us that there is a set of forecasting problems that are common, and increasingly painful, across nearly all finance teams.
On paper, achieving an accurate cash forecast seems straightforward. Simply gather data from across the business, and then collate, harmonize, and present the data in a coherent manner to the CEO, board, or street. Herein lies the problem. As finance professionals know from experience, calculating accurate cash flow forecasts is both challenging and time-consuming. A lack of standardized data and an over-reliance on manual effort can raise questions of forecast integrity.
Difficulties with Data Collation
Accurate cash flow forecasting relies on finance having timely access to multiple data points. But in order to access this data, finance professionals need to overcome multiple hurdles:
- Data collation is typically time-consuming. Data are sourced from many different systems, both internal (CRM, accounting) and external (bank cash management platform). Without an automated feed from these systems, forecasters need to capture the required data manually - an activity that takes time and is error-prone.
- The status of the data varies. Forecasts rely on a combination of certain, predictable, and potential data. To maximize the accuracy of each forecast, the forecaster has to understand the status of all data being used and that this status changes (for example, when a forecast customer payment is received).
- Data may be double-counted, or missed, because different systems are used to source the data. For example, a sale recorded in the CRM system (a predictable cash flow) should result in a cash receipt at the bank (an actual, or certain, cash flow). But if the cash receipt is not efficiently reconciled with the CRM, one sale may be included in the forecast twice. On the other hand, if the sales or procurement teams are slow to update their systems, future cash flows may not feature in the forecast at all.
Difficulties with Data Processing
Collating data is only the first step towards developing a meaningful forecast. Before the forecast can be presented, the data needs to be processed – an often tortuous procedure, for two reasons:
- Data lacks standardization. Because data comes from multiple sources, they may need to be translated into a consistent format before the figures can be manipulated and analyzed.
- Forecasting tools can cause errors. Many organizations still find it easier to forecast from a spreadsheet. But links can break and equations accidentally amended, resulting in inaccurate forecasts. And, if there are weaknesses in the spreadsheet, it will be difficult to identify the reasons behind any variances between forecast and actual positions, rendering future forecasts ineffective too.
Problems lead to Outdated Forecasts
Collating and processing aggregated historical data to produce ‘timely’ cash flow forecasts can be painful and take time. Frustratingly for the finance professional, this means the cash forecast can be out of date even before it is calculated. Outdated forecasts lead to suboptimal decision-making and, in organizations where knowledge of cash is critical to business development, poor forecasting can be disastrous.
There is an AI Way
Yet, companies already have the data they need to create accurate forecasts. The difficulty, historically, has been in accessing the data in real-time and so develop forecasts that enable companies to act strategically on up-to-date information. The answer is to leverage artificial intelligence to improve cash flow forecasts. By automating access to the data across all arms of the business, and providing greater visibility and transparency of data, companies will be better placed to anticipate and address issues as they arise.
Nor are the benefits of automated, real-time forecasts restricted to being a decision-making tool for companies themselves alone. By anticipating and addressing potential issues with customers and vendors before they become problems, relationships along the supply chain can be improved too.