7 Methods of Forecasting Future Market Demands
Forecasting is estimating future demand by anticipating what buyers are likely to do under a given set of conditions. Forecasting is a difficult task for most products or services. Forecasting is relatively easier for products with steady sales or growth in a stable competitive situation.
But most markets lack stable total and company demand. Therefore, accurate forecasting becomes an important factor in company success.
Weak forecasting may result in large inventories, costly price markdowns, or lost sales. The more unstable the demand, the more accurate and elaborate forecasting is needed.
Companies generally follow a three-stage procedure to make a sales forecast. Companies first make an environmental forecast which is followed by an industry forecast. Finally, the company makes its own sales forecast.
The environmental forecast involves projecting inflation, unemployment, interest rates, consumer spending and saving, business investment, government expenditures, net exports, and other environmental events important to the company.
Environmental forecast leads to a gross domestic product forecast, which is used along with other indicators to forecast industry sales. Finally, the company prepares its sales forecast, assuming it will win a certain share of industry sales.
Companies use several specific techniques to forecast their sales.
7 Methods of Forecasting Future Market Demands
|BASED ON :||METHODS|
|What people say||1. Surveys of buyers’ intentions|
2. Composite sales force opinions
3. Expert opinion
|What people do||4. Test markets|
|What people have done||5. Time-series analysis|
6. Leading indicators
7. Statistical demand analysis
This table lists many of these techniques. All forecasts are built on one of three information bases: what people say, do, and have done.
The first basis – what people say- involves surveying the opinions of buyers or those close to them, such as salespeople or outside experts. It includes three methods: surveys of buyer intentions, composites of salesforce opinions, and expert opinions.
Building a forecast on people’s actions involves putting the product into a test market to assess buyer response.
The final basis – what people have done involves analyzing past buying behavior records or using time-series or statistical demand analysis.
Table of Contents
Survey of Buyers’ Intentions
Buyers’ intentions can be known by surveying buyers. Surveys are worthwhile if;
- the buyers have clearly formed intentions,
- will carry them out, and
- can describe them to interviewers.
However, this is not always the case. So, to make forecasts by using consumer survey data, marketers must exercise care.
Various research organizations conduct periodic surveys of consumer buying intentions by using the purchase probability scale.
In addition, surveys can be conducted to know about the consumer’s present and future personal finances and his or her anticipation about the economy.
The various bits of information are combined into a consumer sentiment measure which has been developed by the Survey Research Center of the University of Michigan, USA, or a consumer confidence measure which has been developed by Sindlinger and Company, USA. To make precise forecasts, consumer durable goods companies subscribe to these indexes.
These indexes help them anticipate major shifts in consumer buying intentions so that they can adjust their production and marketing plans accordingly. For business buying, various agencies conduct intention surveys about the plant, equipment, and material purchases.
Composite Sales Force Opinions
Buyer interviewing may not always be practical. In such a situation, the company may make forecasts based on information provided by the salesforce.
The company asks its salespeople to estimate sales by product category for their individual territories. Individual estimates are then added up to make an overall sales forecast. Salesforce’s estimates are used with adjustments for a number of reasons.
Salespeople may be biased in their observations. They may be pessimistic or optimistic. They may go to one extreme or another because of recent sales successes or failures.
Moreover, they may often be ignorant about larger economic developments. They do not always understand the impact of their companies’ marketing plans on the future sales in their territories. They may sometimes understate demand so that the company will set a low sales quota.
If these biases can be checked, a company can gain a number of benefits by involving the sales force in forecasting. Salespeople may have better insights into trends of changes in the market.
Being participants in the forecasting process, the salespeople become more confident in their quotas and get more incentive to achieve them.
Finally, such a “grassroots” approach to forecasting provides the company with estimates broken down by product, territory, customer, and salesperson.
Companies can also take the help of experts in making forecasts. These experts include dealers, distributors, suppliers, marketing consultants, and trade associations. Thus, manufacturers of compact disc players survey their dealers periodically for their forecasts of short-term demand. Dealer estimates have the same strengths and weaknesses as those of salesforce estimates.
Companies can also buy economic and industry forecasts from market research firms. These firms specialize in forecasting and are in a better position than the company to prepare economic forecasts because they have more data available and more forecasting expertise.
Companies may sometimes invite experts to prepare a forecast. The experts may be asked to exchange opinions and arrive at a group estimate. This is called group discussion method.
Or they may be asked to make their individual estimates which are combined into a single estimate by the company analyst. Finally, they may provide individual estimates and assumptions. These are then reviewed by a company analyst, revised and followed by further rounds of estimation.
This is called Delphi method.
Expert opinion can be very useful in making forecasts. But experts may not always be right. For example, in 1943, IBM Chairman Thomas J. Watson predicted, “I think there’s a world market for about five computers”. And in 1946, Daryl F. Zanuck, head of 20th Century Fox, made this pronouncement: “TV won’t be able to hold on to any market it
captures after the first six months”. Where possible, the company should back up experts’ opinions with estimates obtained using other methods.15
The company resorts to test marketing in situations where buyers do not plan their purchases carefully or where experts are not available or reliable. Test marketing is the stage of new-product development where the product and marketing program are tested in more realistic market settings.
A direct test market gives excellent results in forecasting new-product sales or established-product sales in a new distribution channel or territory.
Market forecasting can also be made on the basis of past sales. It is assumed that the causes of past sales can be determined through statistical analysis. This causal relation can then be used to predict future sales.
The statistical tool used for this purpose is called ‘Time-series analysis’. Time series analysis involves breaking down the original sales into four components; trend, cycle, season and erratic components. Analysts then recombine these components to produce the sales forecast.
The trend is the long-term pattern of growth or declines in sales that occur as a result of fundamental changes in population, capital formation, and technology. The trend can be observed by fitting a straight or curved line through past sales against time.
The cycle shows the medium-term wavelike movement of sales due to changes in general economic and competitive activity.
The cyclical component is used for medium-range forecasting. Cyclical changes are difficult to predict because they do not happen regularly. Season indicates a consistent pattern of sales movements within the year. The term season refers to any recurrent hourly, weekly, monthly, or quarterly sales pattern.
The seasonal component may be linked with weather, holidays, and trade customers. The seasonal pattern is useful in forecasting short-range sales.
Finally, erratic components consist of fads, strikes, snow storms, earthquakes, riots, fires, and other disturbances. These components are unpredictable and should be isolated from past data to find the more usual behavior of sales.
Suppose an insurance company sold 12,000 new life insurance policies this year and wants to predict next year’s December sales.
The long-term trend shows a 5 percent sales growth rate per year. This information alone suggests sales next year of 12,600 (=12,000 x 1.05). However, a business recession is expected next year and probably will result in total sales achieving only 90 percent of the expected trend-adjusted sales.
Sales next year will more likely be 12,600 x .90 = 11,340. If sales were the same each month, monthly sales would be 11,340 12 = 945.
However, December is an above-average month for insurance policy sales, with a seasonal index standing at 1.30.
Therefore, December sales may be as high as 945 1.3 = 1,228.50. The company expects no erratic events, such as strikes or new insurance regulations. Thus, it estimates new policy sales next December at 1,228.50 policies.16
Many companies make sales forecasting by using one or more leading indicators. These include other time series that change in the same direction but earlier than company sales.
For example, a sanitary wares manufacturing company might see that its sales lag behind the housing starts index by about three months. The housing starts index would then be an effective indicator.
Statistical Demand Analysis
Time-series analysis considers past and future sales as a function of time rather than as a function of any real demand factors.
But many real factors influence sales. Statistical demand analysis refers to a set of statistical procedures used to identify the most important real factors which affect sales and their relative influence. The factors most widely used are prices, income, population, and promotion.
Statistical demand analysis aims at expressing sales as a dependent variable which is denoted by Q. It tries to explain sales as a function of a number of independent demand variables.
X1, X2, ……., Xn
Thus; Q = f (X1, X2, ……., Xn)
Using a technique called multiple-regression analysis, different forms of the equation can be statistically fitted to the data in the lookout for the best-predicting factors and equation.
For example, a soft-drink company found that the per capita sales of soft drinks by the state was well explained by;
Q = – 145.5 + 6.46X1 – 2.37X2
X1 = mean annual temperature of the state (Fahrenheit)
X2 = annual per capita income in the state (in hundreds)
For example, New Jersey had a mean annual temperature of 54 degrees and an annual per capita income of 24 (in the hundreds). Using the equation, we would predict per capita soft-drink consumption in New Jersey to be
Q = – 145.5 + 6.46 (54) – 2.37 (24) = 146.6
Actual per capita consumption was 143. If the equation predicted this well for other sales, it would serve as a useful forecasting tool.
Marketing management would predict next year’s mean temperature and per capita income for each state and use the equation to predict next year’s sales.
Statistical demand analysis is often a very complex exercise. So utmost care must be taken in designing, conducting, and interpreting such analysis.
The good news is that continuous improvement in computer technology is making statistical demand analysis easier, which is gaining increasing acceptance by marketers as an approach to forecasting.