Understanding Basic Forecasting Terminology | What is Forecasting? – quantitative | qualitative | cycle
Understanding Basic Forecasting Terminology | What is Forecasting?
First, there are two kinds of forecasts, quantitative and qualitative. Quantitative forecasts are based on historical data and statistics. For example, you might look at how much you’ve sold in the past and use that to make a forecast of future sales.
Qualitative forecasts are based on judgment. For example, you might need to predict the sales for a brand-new product where you have no sales history. Or you may know about some upcoming change in taxes or regulations that could change the demand for your product.
Sometimes, you can create a forecast just by looking at a trend. Let’s say that your sales have grown by 10% every year for the last five years. How much would you expect to sell next year? Well, the trend would be 10% more than you sold this year. But things aren’t always linear, and forecasts often need to account for cycles. A common issue is that things can vary throughout the year depending on the season.
So forecasting cycles are often called seasonality. An example of seasonality is that sunscreen sales are higher in the summer than they are in the winter. The further into the future you’re looking, the harder it is to know what will happen. So short-term forecasts tend to be more accurate than long-term forecasts. Now, what makes forecasting so important is that so many decisions are based on our forecasts, but it’s still just a guess. And when you compare the forecast to what really happened, it’s common to find a gap, or a variance. A common way to measure the variance is by calculating how large that gap is in each period as a percentage of the number you were trying to forecast.
This is called the mean average percent error, or the MAPE for short. The lower the MAPE, the more accurate your forecast is. The key to improving your forecasts is to understand what caused your variance. Is it simply a random error, or do you consistently forecast too high or too low? When your forecasts are always high or low, that’s called a bias. And you can correct for bias by changing the way you calculate future forecasts.
Supply chain folks often joke that the first law of forecasting is that the forecast is always wrong, but by understanding the basic concepts and focusing on improving our forecasts, we can help the business make better decisions.
So in the end, forecasting is at the heart of lowering costs, increasing sales, and growing our profits.