When an unexpected event like the pandemic occurs, the initial reaction from management and CFO is to preserve cash and focus on liquidity. The pandemic showed, once again, just how important cash is for businesses.
Of course, it’s hard to predict global disruptions like the recent COVID-19 crisis, and we all hope a similar event might not happen anytime soon. But each company has its own “extraordinary times” and hopefully we all learned how crucial it is to keep tabs on cash. Implementing continuous cash flow forecasting gives you a better overview of cash and some wiggle room to make necessary changes on time.
In short, a cash flow forecast projects the amount of money your company will pay and receive in a set period. Typically it consists of three basic elements: cash balance, cash inflows, and cash outflows. Below you’ll find a breakdown of the process and key steps to implement it in your company.
Where do you start?
Historic data are your starting point for building your cash flow forecast and a better understanding of your business. They are crucial for detecting trends, performing variance analysis, and improving your forecasts with time. The data can come from various sources. Depending on the goal of your cash flow forecasting process and the size of your company you can use bank data (bank statements), ERP or accounting software data. If your data is detailed (transactional) you will need to do the initial grouping of data and mapping it to your cash flow forecasting categories.
Choose a time frame
Depending on the goal of your cash flow forecasting you will need to pick a time for your forecast, in general, they can be:
Short-term - usually forecasting up to 1 month ahead with daily or weekly views. Concerns you’re trying to mitigate with this forecast are mostly focused on keeping the lights on i.e. identifying cash gaps.
Medium-term - monthly projections going up to 12 months into the future. They are also important for liquidity and planning loans.
Long-term - forecasting cash flow 12+ months into the future. This forecast is used for strategic investments and long-term funding decisions.
It’s important to note that the accuracy of cash flow forecasting decreases with a longer time period. For example, in the short-term forecast, you’re using already contracted revenue and days receivable to predict the payments. For the medium and long-term forecast, you might use estimated sales revenue from your managers. In this case, you're forecasting on top of a forecast and naturally, the margin of error is greater.
Cash flow forecasting method
There are numerous methods of forecasting cash flow. Statistical methods like moving averages, exponential smoothing or regression analysis, Percent of Sales method using sales as a key driver, Adjusted net income, Receipts and disbursements method, etc. Picking the right one depends on several factors like:
the level of detail of your forecast
source and structure of your actuals
nature of your business
Perform variance analysis
To assess the accuracy of past forecasts and improve future forecasts it’s important to compare your actual and forecasted cash flow. Bigger deviations should be analyzed in more detail to find their cause and to reevaluate the assumptions for the next forecasting period. It’s important to be diligent about this process and with time (and more actual data) your forecasts will become more accurate.
Variance analysis is tied closely with rolling forecasts. While performing variance analysis you are using the actuals to make changes in assumptions for your next forecast. To get a rolling forecast simply add one more forecasting period (e.g. month) to the forecast.
Were you surprised by the huge impact a few big clients have on your cash flow, or how reducing your DSO for just a few days could resolve many of your cash flow problems?
Since many drivers impact your business, it helps to apply the Pareto principle – target 20% of drivers which impact 80% of your results. Once you know what the key drivers are you can build a driver-based cash flow forecasting model. This way you only need to forecast several items and the rest of the forecast will be calculated from the model.
Use scenario planning
Forecasting predicts possible outcomes based only on existing variables in the model. It offers a narrow (usually just one) range of outcomes and it’s useful for business as usual situations.
Once you have your forecast and you know the key drivers of your cash flow it’s useful to run various scenarios and predict their impact on your business. Start with the key drivers (e.g. a big client delaying their payment or canceling all together). Play with the range of the impact and develop different scenarios. What-if analyses are even more important if you’re forecasting in the medium and long-term periods.
“90 percent of respondents indicated using at least three scenarios to support their planning” according to a recent McKinsey survey of CFOs in leading companies.
Chances are Excel will not be able to support your scenario planning exercise. Find a tool that enables you to forecast your cash flow, run multiple scenarios, compare their impact, archive them for future reference and collaborate with your remote team.
There are some challenges when performing cash forecasting, in general, you should pay attention to these key points:
Perform regular error checks to make sure your data is accurate and up to date
Pick a forecasting method suitable for your business
Make sure your reports provide enough analytical capabilities - stakeholders will need to drill down to specific areas and identify what caused deviations
Avoid wasting too much time on manual tasks like data gathering. If you are overgrowing spreadsheets find a software tool to automate the repetitive tasks
Always have in mind unexpected events can happen and drastically change the accuracy of your projections. To mitigate this have several scenarios in place with an action plan ready to roll out.
If you'd like to see how cloud budgeting software can help your company forecast cash flow click here to book a Farseer demo.