How to generate an accurate cashflow forecast | 3 key factors

| 10-12-2019 | treasuryXL | Cashforce |

One would imagine that in a world where smart cities and virtual reality are becoming a part of daily reality, treasury and finance departments would have perfected their cash forecasting by now, giving the CFO a level of confidence in the numbers. Surprisingly, that doesn’t appear to be the case at all – both PwC’s & Deloitte’s Global Benchmarking studies highlighted cash and liquidity risk as the most important treasury challenge to manage.

If you look closer, it’s not difficult to see why: try managing and forecasting the cash flows of a complex internal ecosystem of multiple ERP’s, FX exposure and geographic entities, combined with increased global uncertainty, tax changes, interest rate rises, and regulatory change. Still, having an accurate cash flow forecast and understanding the underlying drivers is essential to a company’s well-being, as it can help you foresee potential problems which may arise in the year ahead. A lot of companies around the world are therefore increasing their efforts when it comes to cash flow forecasting, but with variable results and accuracy.

So what sets good cash forecasting (i.e. accurate and efficient forecasting) apart from bad cash forecasting (i.e. not transparent, inaccurate & time-consuming cash forecasting)?


A lot of Corporate Treasurers are seeking an accurate cash forecast through a delicate combination of well-chosen cash flow drivers & assumptions. But, to what extent do they have a good view of these cash flow drivers? Do they know what is really eating and feeding their cash (more than the typical high-level AR, AP, Treasury flows that your Treasury Management System will consolidate)?

There isn’t a lot of visibility, unfortunately. Why is that? The classic TMS will typically consolidate basic forecasted flows from the different OpCo’s. The problem is that these OpCo’s cash forecasts are already consolidated from the underlying business transactions. This blurs the insight in the real cash flow drivers and gives no assurance whatsoever on the quality of the data.

To build a good forecast, it is important to have clear and error-free access to the underlying business transactions. In a recent PwC study, only 6% of respondents said they made use of the inputs at the transactional level. But thanks to advances in technology, particularly big data analytics, treasurers can have instant access to the details of the underlying cash movements and are given the ability to drill down to the transaction level. In the gif below, you can see what this means in practice.

Suppose you want to know exactly what drives your company’s cash flow in a certain period. The GIF below demonstrates how easy this could be, using the right platform. Via an easy-to-use click-through interface, the user is able to gain insights per month, quarter, week and day including instant access to the transaction level details.


Cash flow forecasting is often associated with a pile of Excel sheets and manual work. Treasurers are forced to turn to Excel to calculate their forecasts, because classic Treasury Management Systems do not offer the required flexibility.

Getting insights into all your OpCo’s cash flow drivers is one thing but combining all these data sources and applying the right logic/rules to generate a good forecast is another. Let’s take the example of applying vendor payment behavior. Intuitively, it makes sense to enrich invoicing & sales order details with data on when vendors actually pay.  Many companies, however, struggle to take this data into account. In general, they haven’t set up the appropriate algorithms to include in their forecasts. Hence, they face inaccurate forecasts and a lot of time is spent explaining (over and over again) why it was inaccurate.

Defining forecasting logic in a smart way is not an easy challenge. Yet, if your goal is to achieve an accurate forecast, a set of smart logic algorithms is invaluable. Again, modern technology proves to be a great asset. Progressive companies are using technology-driven, smart engines to calculate & automate their cash forecasts, taking over the manually intensive work and proposing logic that could improve the forecast in the future.

Above you can see how a smart engine works in practice. Cash flows are projected into the future (blue line) using forecasting logic. The dotted orange line represents a scenario with one or more of the underlying assumptions changed and immediately shows the impact relative to the blue line.


Even if your forecast is no less than a piece of art, it might be underused, or not used at all. To make a real impact, there should be actions retrieved from the forecast results. There is a lot of potential in accurately predicting what might happen in the future and this potential should be translated into value.

There is even more value in considering multiple scenarios by changing some of the underlying assumptions (e.g. changing the day or frequency of your payment runs). When working in Excel or a TMS, changing assumptions might trigger a lot of additional manual work and is unfortunately often avoided. To get the most out of your forecasting process, it makes sense to build multiple forecasts and assess the impact of each of these scenarios on cash optimization. Driving action combined with building multiple scenarios, can transform finance departments into business partners for fueling a company’s growth.

The orange line reflects a scenario, built by the user. These views give her/him an immediate comparison between the current forecast (full blue line) and a different scenario (based on assumptions made by the user). A powerful simulation engine is able to show the impact of different scenarios in a blink. Imagine the power this can bring to a business-driven finance department.

Mark O’Toole heads up the Americas for Cashforce, a big data analytics & TMS technology provider focused on cash management, forecasting and working capital.