The 5 primary methods for sales forecasting

Regardless of the industry, every company has to include sales forecasting into its greater management strategy. This data set helps establish sales goals in ever-shifting economic climate. Yet this data also involves a bit of guess work: As elements like buying signals and profitability change quite rapidly, businesses must be able to pivot their approach throughout the entire sales season.

Given the unpredictability of most markets, executives have developed a few methods to improve sales forecasting abilities. Understanding these methods is a valuable way to ensure sustained sales growth.

1. Executive judgment
Some companies rely on senior-level executives to create sales predictions. This should mean a great deal of collaboration from executives across several different departments. This pooling of market intelligence can often be successful, However, companies also run the risk of shutting out less experienced executives robbing them of experience and the firm of divergent thinking.

2. Exponential smoothing
This technique involves combining the data from previous years to reach a preliminary forecast of for the next sales period. On the one hand, this method does involve much simpler sets of data overall. However, it doesn’t really let you adjust for seasonality, which can drastically shift the figures depending on a variety of events. To properly account for seasonality, you’d have to then implement the Holt-Winters seasonal method.

3. Naive approach
Let’s say you have a person who orders tomato soup for lunch every single Thursday, and he or she has done so for the last two years. Given the track record, you’d assume that he or she do the same each Thursday. That’s effectively how the naive approach works. It’s a decidedly cheap and direct way to predict sales based on the most obvious data available. However, there is a reason it’s called naive. This approach mostly assumes results, doing little to consider the nuances of the market that might alter outcomes. For instance, the same tomato soup aficionado might go on vacation for a few weeks and thus miss his or her usual Thursday lunch.

4. Delphi technique
In ancient Greek traditions, the Delphic oracle – called Pythia – was an expert in reading the future. Modern businesses have their own Delphic approach, relying on a panel of experts to predict sales outcomes. Businesses need to ensure that they recruit a group of people with diverse opinions. Otherwise, the final predictions won’t be nearly as effective in boosting sales. While the Delphi technique does offer opinion by consensus, it’s worth noting there are downsides. Namely, reaching that very consensus can be difficult depending upon the parties involved.

5. Time series analysis
Much like exponential smoothing, time series analysis is all about looking at previous data to get a sense of possible future outcomes. Time series analysis helps identifying patterns, which the forecaster than extrapolates for the next sales period. Within the umbrella of time series analysis, there are three sub-analyses. Trending looks at the larger, more general patterns that occur between each sales period. Seasonal focuses on patterns that occur around holidays or other drivers. Finally, the random factor involves the impact of environmental conditions like storms and floods.