Basic Marketing: A Global Managerial Approach

Perreault−McCarthy: Basic
Marketing: A
Global−Managerial
Approach, 14/e

21. Developing Innovative
    Marketing Plans

Text © The McGraw−Hill

Companies, 2002

Developing Innovative Marketing Plans 615

Generally, a marketing manager doesn’t have to make forecasts for a national

economy or the broad industry. This kind of forecasting—basically trend project-

ing—is a specialty in itself. Such forecasts are available in business and government

publications, and large companies often have their own technical specialists. Man-

agers can use just one source’s forecast or combine several. Unfortunately, however,

the more targeted the marketing manager’s earlier segmenting efforts have been, the

less likely that industry forecasts will match the firm’s product-markets. So managers

have to move directly to estimating potential for their own companies and for their

specific products.

Many methods are used to forecast market potential and sales, but they can all

be grouped into two basic approaches: (1) extending past behavior and (2) pre-

dicting future behavior. The large number of methods may seem confusing at first,

but this variety has an advantage. Forecasts are so important that managers often

develop forecasts in two or three different ways and then compare the differences

before preparing a final forecast.

When we forecast for existing products, we usually have some past data to go on.

The basic approach—called trend extension—extends past experience into the

future. With existing products, for example, the past trend of actual sales may be

extended into the future. See Exhibit 21-4.

Ideally, when extending past sales behavior, we should decide why sales vary. This

is the difficult and time-consuming part of sales forecasting. Usually we can gather

a lot of data about the product or market or about changes in the marketing envi-

ronment. But unless we know the reasonfor past sales variations, it’s hard to predict

in what direction, and by how much, sales will move. Graphing the data and sta-

tistical techniques—including correlation and regression analysis—can be useful

here. (These techniques, which are beyond our scope, are discussed in beginning

statistics courses.)

Once we know why sales vary, we can usually develop a specific forecast. Sales

may be moving directly up as population grows in a specific market segment, for

example. So we can just estimate how population is expected to grow and project

the impact on sales.

The weakness of the trend extension method is that it assumes past conditions

will continue unchanged into the future. In fact, the future isn’t always like the

past. An agent wholesaler’s business may have been on a steady path, but the devel-

opment of the Internet adds a totally new factor. The past trend for the agent’s sales

changed because the agent could quickly reach a broader market.

As another example, for years the trend in sales of disposable diapers moved

closely with the number of new births. However, as the number of women in the

workforce increased and as more women returned to jobs after babies were born, use

of disposable diapers increased, and the trend changed. As in these examples, trend

Two approaches

to forecasting

Extending past

behavior can miss

important turning

points

Dollars

(^0) Years
Actual sales
Trend
Exhibit 21-4
Straight-Line Trend
Projection—Extends Past
Sales into the Future