What’s it: Sales forecasting attempts to predict future sales for a specific period and with certain basic assumptions. It may be a sales value or sales volume. In contrast to sales volume, you need to assume price and volume to predict the sales value. The predictions might be for the next year or the next three to five years.
Importance of sales forecasting
The sales forecast is the basis of planning. Predicting sales is important for production and inventory, budgeting, marketing programs, purchasing, and resource allocation plannings.
Indeed, sales forecasts are often not 100% accurate so as not to reduce future uncertainty. In other words, companies can’t predict the future with accurate certainty.
However, sales forecasts increase management’s confidence to make critical business decisions. They include:
- Managing cash flow and setting budgets for each of the business functions.
- Estimating the need for the supply of raw materials, labor, and production machines.
- Determining the optimum level of inventory so it is not too excessive or too low.
- Developing appropriate marketing plans, including pricing, products, promotions, and distribution channels.
- Estimating human resource needs, for example, whether to recruit permanent or temporary workers.
For example, a sales forecast might require the company to adjust its existing resources. The current production capacity may be insufficient, so they decide to buy a new machine. They need a more aggressive promotional program because future sales are prospective, along with high market demand.
Sales forecasting methods
Two types of sales forecasting:
If you like our curation and click to continue buying, thanks for contributing to us. We may earn a commission when you buy through our links. Learn more ›
- Qualitative, relying on subjective judgments to generate estimates.
- Quantitative, relying on data to generate a prediction.
In this approach, companies use subjective judgments of several people instead of being based on data. They may conduct salesforce polls and consumer surveys. Alternatively, they can use:
- Program evaluation and review technique (PERT)
- Delphi technique
Under the PERT method, the expert or salesperson provides forecasts in three scenarios: optimistic, pessimistic, and most likely. Then, they weighted each scenario with a certain weight to determine the final result. For example, suppose they think a most likely scenario has double the odds of a pessimistic and optimistic scenario. In that case, we can calculate the PERT estimate as follows:
PERT estimate = (Optimistic + (2 x Likely) + Pessimistic) / 4
The formula above also assumes the optimistic scenario has an equal probability as the pessimistic scenario.
Furthermore, under the Delphi technique, the company gathers a panel, usually composed of external experts. Each produces independent forecasts, which then revise their projections until they reach consensus.
The qualitative method has several advantages. Some companies use them because they are relatively simple, especially if historical data are unavailable. Also, this method does not rely on quantitative data but on the experience of experts. Companies love it, significantly when the determinants of buying habits change or are difficult to determine.
Qualitative methods rely on data to generate estimates. In general, quantitative methods use past sales data to predict future sales. We call that time series analysis. Some of the techniques are Exponential Smoothing, Autoregressive Model, Moving Average, and Autoregressive Integrated Moving Average (ARIMA).
Meanwhile, another technique is to rely on several predictor variables to generate sales forecast figures. Say, a company uses economic growth and expenditure per capita of the population as predictors. The most common analysis is regression. It might use one point in time.
A more sophisticated regression model is panel regression. It uses multiple points in time (panel data) and multiple variables. For example, to predict sales, a company uses economic growth data and the population’s per capita expenditure in the last ten years.
Factors to consider in forecasting sales
Forecasting helps us to provide a picture of the future, which is at least close to it. However, it is impossible to come up with exact figures because the future is uncertain. Various external factors affect sales performance, including:
- Changes in consumer tastes
- Technology changes
- Changes in shopping behavior
- Economic conditions such as economic growth, inflation, exchange rates, and interest rates
- Competition dynamics, including competition strategy