Clearly, the better management’s ability to predict future internal and external environmental condition in a particular situation the better its chances of formulation attainable plans. Infact, early management writer like Henri Fayol considered predicting the future and planning accordingly to be the very essence of management. In his view, all organizational plans were an amalgamation of several forecasts.

Forecasting is a technique that uses both past experience and present assumptions about the future to predict what will occur. When forecasting is properly performed, the resulting forecast is a vision of the future that can reasonably be used as a premise for planning. Forecasting today is a specialized field with several sub-specialists. There are companies whose only business in making forecasts in a specific area. An example is the GALLUP organization in the USA which specializes in gathering and analyzing information that enables, one to make predictions bout preferences and outcomes various political and social issues, the objective of forecasting is non to gain certainty; it is to reduce uncertainty. A second benefit of forecasting is that the process may well begin for managers, the process of questioning these managers to forces and changes that will continually be affecting the future success of the organizational strategy for these reasons, forecasting is utilized by most managers in some form, even though it may be informal or only an initiative process. thus, manager who male a snap decision not to replace a piece of worn out machinery is implicitly forecasting that his profit will be higher than if he bought a new one. This sort of initiative approach to managing can be successful as often as not unforeseen consequences occur and create problems for the manager


Before the industrial revolution the time cycle between product conception and introduction could be very short. When a new product is conceived, an entrepreneur fabricates it in his new workshop admits cheap labour, and present it to an untapped market within months. The product was usually built for long life and had a long cycle of exploitation. New, the situation is different. Products tend to be very much more complex, long development times and escalating cost are major features, because of, the rate of technical change is such that thy tend to become obsolescent more rapidly, and the profit recovery must be made over less sales volume and in shorter time. efforts must continually be made to shorten development times to maximize the product’s saleable life before obsolescence. In areas like chemical processes involving synthetics- it is possible to go from innovation to full commercial exploitation in one or two years whereas in other areas, for example, military equipment a ten year cycle is not unusual. There is thus a prime need to forecast future technical and market trends. Extrapolation can be dangerously misleading. For instance, companies building steam locomotive were left high and dry when railways were nationalized to form British railway which decided to move from s team to diesel and electric locomotives. Many had to diversify, rapidly into a variety of products in which they had no experience, sometimes with disastrous result. It is the duty of long –range corporate planners to maximizes future opportunities and minimizes future risks by identifying and evaluating probable trends and alternative methods of utilizing resources.


Technological forecasting is concerned with the application of scientific methods of the analysis of technical trends, taking full consideration of the effect of possible changes in the political and social environment.

It has made tremendous progress in the last decade and although the basic principles are simple, the techniques used are becoming increasingly precise facilitated by the availability of high-power computers.

To appreciate the problems of forecasting, it is necessary to know something about the development of technology and human attitudes to change. By virtue of this genetic make up, including such attributes as manual dexterity, stereoscopic vision and the ability to reason, man is inherently inventive.

Again, we are unable to recognize changes in the features and characteristics of relatives and colleagues unless the change is dramatic because of illness or accentuated by a period of separation of some weeks or even months. Thus, we are unable to recognize changes in the total environment, unless special techniques are established to select such change, and so we are always living in an environment that has ceased to exist. Environment is like yeast, constantly working to change the material elements in which it is in contact.


Any manager who has recently been concerned with the application of forecasting in his decision making is well aware of the importance of selecting the appropriate set of guidelines that can be used in matching them with specific situations are not yet been developed because of the newness of these methods.

Before a forecast is made, six major characteristics deserve mention

The time Horizon: the duration with which a decision will have an impact and for which a manager must plan clearly, affect the selection of the most appropriate forecasting technique. Such as –

Immediate term (less than one month).

Short term – one month to three months

Medium term – three months to two years

Long term – more than two years

Level of Detail: The normal decision making task in most corporations are subdivided for easy handling according to the level of detail required, e.g. having a planning department that does the aggregate planning, perhaps by product group or the entire sales of the company.

Number of Items: When decision is made concerning hundred or even thousands of product companies have found it most effective to develop simple decision rule that can be applied mechanically to each of the items. The same general principles holds true in forecasting.

Others are: Control versus Planning, Stability, Existing Planning procedures etc.


The Time Horizon – two aspects of the time horizon relate to individual forecasting methods. First, is the apan of time into the future for which different forecasting methods are best suited. Generally speaking, QUALITATIVE methods of forecasting are used much more for longer term forecasts, where as QUANTITATIVE methods are more appropriate to the intermediate and short term.

The second important aspect of time Horizon is the number of periods for which a forecast is desired. Some techniques are appropriate only for forecasting one or two periods in advance, but other techniques can be used for several periods into the future.

The Pattern of Data – underlying the majority of forecasting methods is the assumption of the type of pattern found in the data to be forecast; for example, some series depict a seasonal as well as a trend pattern. Other may simply consist of an average value with random fluctuations surrounding it. Because different patterns, it is important to match the presumed pattern in the data with the appropriate technique.

Type of Models – in addition to pattern of data, most forecasting method also assures some model of the situation being forecasted. This model may be a series in which time is viewed as the important element to determine change in the pattern or it may be statistical in nature – regression or correlation analysis. Other such as casual models that represent the forecast as being dependent on the occurrence of a number of different events or mixed in which a number of different events or mixed in which a number of different models are actually combined, are also available. The understanding of the mathematics is not important to the decision maker but the assumption underlying each model are different and the model capabilities are different to different decision making situations.

Cost – Generally four elements of cost are involved in the application of a forecasting procedure, development, storage, actual operation and opportunity in terms of other techniques that might have applied. The variation in cost obviously has an on the attractiveness of different methods for different situations.

Accuracy – closely related to the level of detail required an a forecast is the accuracy required. For some decision makers anywhere between plus or minus 10% may be sufficient for their purposes, but in other cases a variation of as much as 5% could spell disaster for the company.

Ease of Application – One general principle in the application of scientific methods to management is that only those methods that are understood are actually used by the decision maker.


There are many forecasting methods. Although these methods are different, they should not necessarily be regarded as mutually exclusive. Indeed, some methods may be more suitable for preparing short term forecast such as monthly or quarterly predictions; others may be best for long term projections of a year or more. Some may be preferred for forecasting at the level of the firm.

MECHANICAL EXTRAPOLATIONS: Extrapolation procedures of one form or another are extensively used by business executives, economists, market research and others engaged in forecasting activities. As a method of predicting, extrapolation may include procedures ranging from simple coin tossing to determine fluctuation to the projection of trends, autocorrelations, and other seemingly more complex mathematical techniques.

Also See:  Popular Trade Associations In Nigeria

TIME SERIES ANALYSIS: A time series is a set of observations taken at specific times, usually at equal intervals. It is a sequence of values corresponding to particular points or periods of time. Data such as sales, production and prices when arranged chronologically, are thereby ordered in time and hence are referred to as time series. Example of time series are the total Gross Domestic Product of Nigeria over a number of years; the total monthly sales receipts of a department store and the daily closing price of a share on the stock exchange. Managers who use time series forecasting are banking on past trends continuing unchanged into the future.

There are at least four sources of variation at work in a time series;

i. Trend (T)

ii. Seasonal variation (S)

iii. Cyclical variation (C)

iv. Irregular forces (I)

TREND – represents the long-term growth or decline of the series.

SEASONAL VARIATIONS – due to weather and custom manifest themselves during the same approximate time periods each year (for example, Christmas, Eid-el-kabir, Easter and other season of the year during which different types of purchases are made).

CYCLICAL VARIATIONS – These cover several years at a time, reflect prosperities and Recessions. And finally,

IRREGULAR FORCES – such as strikes (ASUU) Academic Staff Union, war and boycotts, are erratic in their influence on the particular series, but nevertheless must be recognized.

Of the four forces affecting time series technique, the seasonal factor is fairly easy to measure and predict. The irregular factor is unpredictable but can be adjusted by smoothing out process such as moving average. Hence, the trend which represents persistent growth or decline, and the cyclical which is presumably recurrent, are the forces which have occupied the chief attention of forecasters using time-series analysis.


When time series is properly utilized, there are a number of advantages which can be derived from the method. Some of the advantages are as follows:

The necessary data are usually minimal and often easily obtained either form within the company itself or from readily available outside sources.

The analytical calculations, such as the moving average are usually simple and repetitive, and therefore suitable for computer processing. Hence, these techniques may be particularly well suited for problems in which a large number of variables must be forecast.

Only moderate analytical skills may be required of the forecasters themselves. The methods are fairly easy to understand and the data processing straight forward.

The method is largely objective, although judgement is involved in choosing additive or multiplicative, decomposition, fixed or changing seasonal factors, type of trend to use and extrapolation of the cyclical component.

The resultant forecasts are usually reasonably accurate for the short run, say, a twelve month period.

Time series analysis usually permit the calculation of the degree of error in the forecast. Hence an interval of confidence attendance with the predicted value strengthen the quality of the forecast itself. Forecast errors can be further reduced where identification of dependable trend and seasonal patterns can be made.

Once the composition of the time series has been accomplished the way is open for a casual analysis of the separate components.


Time series analysis cannot be used in situations where time series data have not been accumulated for example, projections for a new product or new environment for which no historical records have been kept.

Forecast based on extrapolation of trend, cyclical and seasonal component of a series assume a strong persistence of time patterns from the past into the future. This may not always be a valid assumption.

Strict adherence to the time series analysis techniques will fail to take advantage of the forecasters knowledge of prospective development. For example, the forecaster might know what advertising effort will be greater than anything in the past, and such knowledge should be used to modify the extrapolation.

Time series analysis gives no information as to casual factors influencing the time-series components. It merely provides a basis for causal analysis.


Causal forecasting techniques help show a manager why his sales are tapering off or why they are higher every two years. They are used when enough analysis has been performed so that clear relationship between the company factor (such as sales) and other factors (such as Gross Domestic Product) can be ascertained.

Basically causal forecasting involves estimating the company factor (such as sales) from the other factors (such as cost of electricity), (or advertising expenditures) usually, “correlation analysis” (how closely are the variable related) and “regression analysis” (how can one factor be predicted if the values of related factors are known?) are used in order to develop the necessary relationship.

Although time series forecasting and causal forecasting can be very useful, they clearly have some limitations. First, they are virtually useless when data are scarce – such as for an entirely new product which has no sales history.

Secondly, they assume that historical trend will continue into the future. But trends have a habit of changing, often suddenly; as a British economist once pointed out, “A trend is a trend. The question is when will it blend? Will it climb higher and higher or eventually expire, and come to an untimely end”.

A third limitation is that both of these forecasting techniques tend to disregard unforeseeable, unexpected occurrences. And yet it is these unexpected occurrences (such as the fuel or bread shortage) that often have the most profound effect on companies.


Qualitative forecasting techniques are used when historical data is scarce and when you are especially interested in identifying unexpected future opportunities or threats. These techniques emphasise human judgement. They gather in as logical, unbiased and systematic a way as possible, all the information and judgement that can be brought to bear on the factors being forecast. Two such techniques are Market Research and the Delphi Method.


MARKET RESEARCH – Market research is the systematic gathering recording, and analyzing of data about problems related to the marketing of goods and services. It is aimed at obtaining answer to questions such as “will such product appeal to a particular type of customer, “will the price be an insurmountable disadvantage”? “will it be a bigger seller in a particular part of the country?” It is used to obtain information concerning consumer buyer preferences, effectiveness of advertising and so forth. Typically data are collected through questionnaires, direct observation and analysis of secondary” sources of data such as census reports.


DELPHI FORECASTING TECHNIQUES – The Delphi technique is often used in order to make technological forecasts. This technique involves obtaining the opinions of people who are experts on the future technological and economic trends which might affect company’s market. But instead of putting all these experts together in one room, their opinions are solicited anonymously and individually through questionnaires. These opinions are analysed, distilled and resubmitted to the experts for a second round of opinions. This process may continue for five, six or more time.




For an effective forecasting, there has to be some way of measuring its performance. Having one’s forecast checked against actual results and against the forecast and opinion of others can be a disenchanting experience. For organization own protection, as well as to encourage intelligent use of their products, economic forecasters should insist upon objective and systematic procedures for review and rigorous evaluation of their work.

These procedures must recognize that forecasting with complete accuracy is impossible and fortunately unnecessary, provided the forecasts are evaluated and use in the proper manner. Even if the forecaster’s estimate of some economic indicator is pretty far from work, the correct operating decision can be reached if the forecaster is generally correct in the appraisal made of the other forces at work in the economy.




i. After the forecasted event has happened or not happened, the accuracy of the forecast to those who may have acted upon it should be appraised.


ii. Between the time the forecast is released and the time of the forecast events, an interim review should be made to determine whether or not to revise, the forecast.


iii. While the forecast is being prepared or just prior to its release, an evaluation of its quality should be made in full recognition that important and perhaps costly decision might be based upon it.

Related Posts

Marketing Information System (MKIS) & Marketing Research(Opens in a new browser tab)

The Marketing Role Facilitators(Opens in a new browser tab)

Objectives in Management

How To Start Home Based Business(Opens in a new browser tab)

Market and Marketing

Project Background

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Click Here To Call Us