There are two kinds of tools that one can use to estimate market demand. The qualitative (mainly survey) and quantitative. Let us examine these tools in greater detail.
(1) Qualitative Tools
Qualitative tools involve opinion surveys. Some of the most prominently used ones are described below:
â€¢Survey of Buying intention
This involves surveying the buyers to assess their intentions to buy the product. This is very useful in estimating the market demand for customer durables. Or even in case of new products. This method can also be used to measure the demand of a product at different levels of marketing effort. For example, change in price, and its effect over the consumers demand can be studied through this method. The purchase intentions of the buyer can be measured on a seven point scale from â€œdefinitely buy â€œand definitely not buy â€œ.The response so obtained constitutes purchase probability for a given product and hence an index of purchase probability can be made .This method is also useful in industrial marketing Though it is a useful method, it suffers from the same limitations as any other consumer survey does, as described earlier.
â€¢Composite of Sales Force Opinion
In this method, the company asks individual sales personnel to estimate the sales of the given product in his or her territory. These estimates are then pooled and a national level forecast of sales is obtained .Very few companies use this tool because, quite often, sales people are believed to under estimate sales in their territories. The reason is that they would like to show a positive variance of sales against targets to their top management. It is for this reason that not many companies rely on sales force opinion poll.
This involves constituting a panel of experts and asking them to estimate the market demand for a given product. They are also asked to mention their assumption about the future market environment .Individual experts do not know who else is on the panel .Since each expert works from his or her office, the chances of him or her getting influenced by others does not arise .Once the marketer gets the estimates ,he or she isolates extreme opinions and estimate and reverts back to the concerned expert giving them the assumptions which others have made However, the marketer does not reveal the estimate of the other experts. The objective of sending back extreme opinions is to get a consensus. But, should the extreme opinion holders choose not to revise their opinions the marketer will have to leave it at that. But this method can help to know the different scenarios and is particularly useful in estimating demand for a new product or technology.
A variant of the Delphi Technique is the expert opinion poll in which a firm may interview experts in its industry. These experts could be dealers, large buyers; marketing consultants and trade associations .But these polls have the same limitations as that of the consumer survey. Nevertheless, these polls are commonly used by many firms for estimating the market demand and the companyâ€™s market share.
The quantitative techniques could further be categorized as:
1.Tools for short-term forecasting.
2.Tools for long-term forecasting.
Short â€“term Forecasting
The short-term forecast refers to all forecasts up to a period of one year. Most often the sales managers are interested in this forecast. Tools commonly used here are clubbed as extrapolation techniques .Illustration of these are exponential smoothing, times series decomposition and several naive models. The most common is exponential smoothing technique which is a type of moving average that represents a weighted sum of all past numbers in the time series with the heaviest weight placed on the most recent information .This method involves estimating the value of â€œsmoothing constant â€œ ( usually designated by the symbol a) and then using it to â€œsmoothâ€? the raw sales data .The assumption in this method is that the actual sales is a function of environmental factors and the method helps to â€œsmoothâ€? out these factors .The exponential smoothing method can be represented as being given below :
St= a Xt (1-a) St -1
Where â€œSt â€œ refers to smoothed sales in period t
â€œaâ€? is smoothing constant with a value between 0-1.
â€œXtâ€? is actual sales in period t
â€œSt-1 â€œ is smoothed sales in period t-1
A major challenge is that of estimating â€œaâ€? value .In fact, this problem of assigning a value to â€œaâ€? creates a limitation in the usage of this method. A general principle used here is that so the time series changes very slowly the value of â€œaâ€? could be small to keep the effect of earlier observations. But if the changes are too rapid â€œaâ€? value will have to be high to give forecasts responsive to these market changes .In reality, the value of â€œaâ€? is estimated by trying several values and making retrospective tests of the associated error function. The â€œaâ€? value leading to the smallest error is then chosen for future smoothing.