William C. House, in describing the Insect Control Services Company, developed an excellent illustration of how forecasting works. In general, Insect Control Services forecasts by attempting to do the following:⁴³

1. Establish relationships between industry sales and national economic and social indicators.⁴⁴

2. Determine the impact that government restrictions on the use of chemical pesticides will have on the growth of chemical, biological, and electromagnetic energy pest-control markets.

3. Evaluate sales growth potential, profitability, resources required, and risks involved in each of its market areas (commercial, industrial, institutional, governmental, and residential).

4. Evaluate the potential for expansion of marketing efforts in geographical areas of the United States as well as in foreign countries.

5. Determine the likelihood of technological breakthroughs that would make existing product lines obsolete.

In addition to the general type of organizational forecasting done by Insect Control Services, specialized types of forecasting, such as economic, technological, social trends, and sales, are available. Although a complete organizational forecasting process should, and usually does, include all these types of forecasting, sales forecasting is considered the key organizational forecast. A sales forecast is a prediction of how high or low sales of the organization’s products or services will be over the period of time under consideration. It is the key forecast for organizations because it serves as the fundamental guideline for planning. Only after the sales forecast has been completed can managers decide, for example, whether more salespeople should be hired, whether more money for plant expansion must be borrowed, or whether layoffs and cutbacks in certain areas are necessary. Managers must continually monitor forecasting methods to improve them and to reformulate plans based on inaccurate forecasts.⁴⁶

Modern managers have several different methods available for forecasting sales. The two broad types of sales forecasting methods are qualitative and quantitative. In the following sections, we highlight popular qualitative (i.e., jury of executive opinion, salesforce estimation) and quantitative (i.e., moving average, regression, product stages) forecasting methods.

Quantitative Methods

moving average

The moving average method utilizes historical data to predict future sales levels. Specifically, forecasters compute average sales levels for x historical time periods; forecasters are able to choose the time periods that best fit their situations. Suppose, for example, that forecasters at Toyota are using a five-year moving average to predict future automobile sales. In 2015, they would select the five most recent years—2010 to 2014—and compute average automobile sales during that period. In 2014, they would have relied on sales data from 2009 to 2013, and in 2013, they would have relied on sales data from 2010 to 2014. Because the five-year time period changes each year to reflect the five most recent years, this method is referred to as a “moving” average.

regression

The regression method predicts future sales by analyzing the historical relationship between sales and time.⁵⁰ Using this information, analysts can use regression to forecast future sales. Specifically, regression provides forecasters with a trend line that best illustrates the historical relationship between sales and time. Forecasters can use this trend line, then, to predict future sales.Figure 5.7 illustrates an example of a trend line that can be used to forecast future sales. Managers often use statistical programs such as SPSS or SAS to conduct regression analysis.

Although the actual number of time periods included in regression will vary from company to company, as a general rule, managers should include as many time periods as necessary to ensure that important sales trends do not go undetected. For example, at the Coca-Cola Company, management believes that to validly predict the annual sales of any one year, it must chart annual sales in each of the 10 previous years.⁵¹

product stages

The data in Figure 5.7 indicate steadily increasing sales for B. J.’s Men’s Clothing over time. However, because in the long term products generally go through what is called a product life cycle, the predicted increase based on the last decade of sales should probably be considered overly optimistic. A product life cycle is made up of the five stages through which most products and services pass. These stages are introduction, growth, maturity, saturation, and decline.⁵² The product stages method predicts future sales by using the product life cycle to better understand the history and future of a product.

Figure 5.8 shows how the five stages of the product life cycle are related to sales volume for seven products over a period of time. In the introduction stage, when a product is brand new, sales are just beginning to build (e.g., ultra high-definition televisions). In the growth stage, the product has been in the marketplace for some time and is becoming more accepted, so product sales continue to climb (e.g., smartphones, tablet computers). During the maturity stage, competitors enter the market, and although sales are still climbing, they are climbing at a slower rate than they did in the growth stage (e.g., personal computers). After the maturity stage is the saturation stage, when nearly everyone who wanted the product has it (e.g., refrigerators and microwaves). Sales during the saturation stage typically are due to the need to replace a worn-out product or due to population growth. The last product life cycle stage—decline—finds the product being replaced by a competing product (e.g., conventional, or not high-definition, televisions).

Managers may be able to prevent some products from entering the decline stage by improving product quality or by adding innovations. Other products, such as scissors, may never reach this last stage of the product life cycle because there are no competing products to replace them.

The sales forecasting methods just described are not the only ones available to managers. Other, more complex methods include the statistical correlation method and the computer simulation method.⁵³ The methods just discussed, however, do provide a basic foundation for understanding sales forecasting.

In practice, managers find that each sales forecasting method has distinct advantages and distinct disadvantages. Therefore, before deciding to use a particular sales forecasting method, a manager must carefully weigh these advantages and disadvantages as they relate to the manager’s organization. The best decision may be to use a combination of methods to forecast sales rather than just one method. In fact, recent research suggests that combining quantitative and qualitative forecasting methods, as opposed to using only quantitative or only qualitative methods, results in better forecasts.⁵⁴ Whatever method or methods are finally adopted, the manager should be certain the method is logical, fits the needs of the organization, and can be adapted to changes in the environment.

One study surveyed forecasters to gauge their familiarity with using these forecasting methods.⁵⁵ The authors of the study then compared these familiarity statistics with two similar studies conducted in the 1980s and 1990s. The results of the study, which are displayed in Table 5.2, reveal some interesting trends. First, the results suggest the increasing popularity of quantitative forecasting methods; in fact, 100 percent of forecasters polled in the 2000s were familiar with the moving average method. In contrast, familiarity with qualitative methods—especially the jury of executive opinion method—has decreased over time.

Alternate View

	1980s	1990s	2000s
Qualitative Methods
Jury of Executive Opinion	87%	82%	74%
Salesforce Estimation	84	85	83
Quantitative Methods
Moving Average	92%	98%	100%
Regression	80	88	97

Note: The numbers in this table reflect the percentage of respondents who were “familiar” or “somewhat familiar” with the corresponding forecasting method.

The Gantt chart, a scheduling device developed by Henry L. Gantt, is essentially a bar graph with time on the horizontal axis and the resource to be scheduled on the vertical axis. It is used for scheduling resources, including management system inputs such as human resources and machines.

Figure 5.9 shows a completed Gantt chart for a work period entitled “Workweek 28.” The resources scheduled over the five workdays on this chart were the human resources Wendy Reese and Peter Thomas. During this workweek, both Reese and Thomas were supposed to produce 10 units a day. Note, however, that actual production deviated from planned production. There were days when each of the two workers produced more than 10 units as well as days when each produced fewer than 10 units. Cumulative actual production for workweek 28 shows that Reese produced 40 units and Thomas 45 units over the five days.

The main weakness of the Gantt chart is that it does not contain any information about the interrelationship of tasks to be performed. Although all tasks to be performed are listed on the chart, it is not possible to tell whether one task must be performed before another can be started. The program evaluation and review technique (PERT), a technique that evolved partly from the Gantt chart, is a scheduling tool that does emphasize the interrelationship of tasks.

Critical Path

Managers need to pay close attention to the critical path of a PERT network—the sequence of events and activities requiring the longest period of time to complete. This path is called critical because a delay in completing this sequence results in a delay in completing the entire project. The critical path in Figure 5.10 is indicated by thick arrows; all other paths are indicated by thin arrows. Managers try to control a project by keeping it within the time designated by the critical path. The critical path helps them predict which features of a schedule are becoming unrealistic and provides insights into how those features might be eliminated or modified.⁶¹