Financial forecasting arms executives with specific and accurate predictions they can use to make plans for the company, barring unforeseen circumstances. Those plans can affect everything from budgeting, hiring, sales goals and earnings predictions to financing decisions and institutional investment goals.
In short, financial forecasting is at the very heart of every decision executives make. Without it, they’re blindly leading the company forward and possibly off a cliff.
Financial forecasting is not a one-size-fits-all practice. It’s a collection of techniques and methods that executives choose from depending on the data they’re using and the purpose of the output.
What Is Financial Forecasting?
Financial forecasting refers to a process businesses use to predict future revenues, expenses and cash flow. Executives use financial forecasting to help them make confident, profitable financial decisions and be able to determine where the company is headed.
What Are the 4 Financial Forecasting Methods?
Financial forecasting methods fall into two broad categories: quantitative and qualitative. The first relies on data that can be measured and statistically controlled and rendered. The latter relies on data that cannot be objectively measured.
It’s important to note that no financial forecast is foolproof since you are mapping the road ahead by looking in the rearview mirror. However, when done properly, forecasting is generally reliable.
1. Straight Line Forecasting Method
This method is commonly used when the company’s growth rate is constant, to get a straightforward view of continued growth at the same rate. It involves only basic math and historical data. Ultimately, it renders growth predictions that can guide financial and budget goals.
An example of straight line financial forecasting
A restaurant chain’s annual growth rate has held steady at 5% over the past three years. The company expects its growth to continue at that rate over the next two years. By calculating next year’s growth at 5% over this year’s, and the following year’s at 5% above next year’s, the company can make accurate predictions about how many people it will need to hire and the added payroll costs for each of those years.
2. Moving Average Forecasting Method
A moving average is the calculation of average performance around a given metric in shorter time frames than straight line, such as days, months or quarters. It is not used for longer time periods, such as years, because that creates too much lag for it to be useful in trend following.
This method is used to create a constantly updated average of values with a lot of movement, such as stock prices, as well as values that fluctuate often but not quite as quickly, such as inventory levels during peak retail periods.
In short, this method helps identify underlying patterns which you can then use to evaluate common financial metrics such as revenues, profits, sales growth and stock prices. A rising moving average indicates an uptrend, whereas a falling moving average points to a downtrend.
An example of moving average financial forecasting
A retailer wishes to calculate how much—if any—product he should reorder from a wholesaler. It’s the holiday season, so sales are going well overall, but he needs to know which products are trending upward. Rather than try to track sporadic upticks and drops in a specific product’s sales throughout the day or over a week, he calculates a moving average for the week to show him the trend and drive his inventory purchase orders.
3. Simple Linear Regression Forecasting Method
It is used to chart a trend line based on the relationship between a dependent and independent variable. A linear regression analysis shows the changes in a dependent variable on the Y-axis to the changes in the explanatory variable on the X-axis. The correlation between the X and Y variables creates a graph line, indicating a trend, which generally moves up or down, or holds consistent.
An example of simple linear regression forecasting
Sales and profits are two variables that are key to the success of every company. Using the simple linear regression method, if the trend line for sales (x-axis) and profits (y-axis) rises, then all is well for the company and margins are strong. If the trend line falls because sales are up but profits are down, something is wrong; perhaps there are rising supply costs or narrow margins. However, if sales are down but profits are up, the value of the item is trending upward. This means the company’s expenses/costs are down and the linear regression is good—the margin percentage is up when profits are up.
4. Multiple Linear Regression Forecasting Method
This method uses more than two independent variables to make a projection. Basically, multiple linear regression (MLR) creates a model of the relationship between the independent explanatory variables (parameters) and the dependent response variable (outcome).
An example of multiple linear regression
A trucking company executive wants to predict fuel costs in the next six months. The independent variables she uses for this method are the EIA Gasoline and Diesel Fuel Update, oil futures from a futures exchange, mileage from GPS fleet routing systems, traffic patterns from smart city open data platforms and the number of trucks the company expects to be on the road during the period based on delivery orders. This list is for illustrative purposes only, and other variables may also affect the result (outcome).
In any case, all of the variables are independent of the outcome but also have an effect on the outcome. This model predicts the outcome—in this case, the predicted fuel costs for the period—based on the variables.
A Note on Qualitative Forecasting Methods
By their nature, qualitative forecasting methods are less precise than quantitative. They’re as much art as imprecise science. That is not to say, however, that they are less useful.
For example, a doctor learns from experience the telltale signs of a certain disease, which drive his decision to order certain tests. The doctor may also suspect one disease over another because one is common in the local area, even if uncommon nationally.
Similarly, business executives develop expert knowledge from experience pertaining to their industry or product line. This information is not necessarily measurable, nor confirmed by historical data, but it has business value nonetheless.
Qualitative forecasting methods use or combine soft data, such as expert estimates or opinions, with hard data, such as machine data or sales data, to make projections that are usually applied to short-term business predictions.
One example is the Delphi method, which is similar to market research methods but incorporates soft data from subject matter experts. It might entail the use of questionnaires, rather than data gathered from consumer responses to a product or service.
Choosing the Right Method for Your Business
Financial forecast method selection is based on several considerations, primarily:
- The context of the forecast.
- The relevance of available historical data.
- The purpose of the analysis.
- The window of time in which the analysis must be completed or applied.
Generally speaking, more variables give you a better-defined result since context and other factors are considered in the calculation. However, simpler methods are useful when you just want a straightforward answer—one of these methods would be to opt for financial management software. You can also select methods for one purpose and use their outcomes in another method for another purpose. For example, the results for either or both straight line and moving average can be used as variables in either of the linear regression methods.