Page 119 - Marketing the Basics 2nd

price 111
per cent off if the product is not moving or allow the store to return the merchandise. Armani then gets rid of the goods to close-out stores or other discount channels.
PREDICTING DEMAND
Many models exist to predict demand. Each one measures certain variables which in turn, offer an insight on the dynamics within a market. However, due to the nature of the modelling process, some elements are always ignored. The hope is that the model ignores the inconsequential variables, or more formally stated, minimizes the error while measuring the variance between observations. As such, when estimating which of the eight states of demand your product is experiencing, it is prudent to use more than one model and subject it to many tests. We discuss three types below.
The first type of model is to use past history to predict the future. In its simplest form, the relationship between demand and price is calculated and is used to predict future behaviour. This is often used when coming up with the sales quotes for the next year; if inflation is 3 per cent and the firm wants to grow at 5 per cent, all quotes are raised 8 per cent for next year. If the product is being sold in a mature market, it is plausible a simple model is all that is needed. But since markets are in a constant state of flux due to changing consumer tastes, technological innovation and insatiable growth appetites of corporations, more complicated statistical models on which to form these forces are needed. More often than not, such elaborate models provide better results versus a simplistic model. The key to employing statistics to predict demand is to build a model that is appropriate to the market in which the product exists.
A second approach to conduct price experiments. Varying the price of a product at the same store or at different locations of a franchised chain, assuming all other factors are constant, will yield an accurate portrayal of the demand curve. This can be done in supermarkets where they have many stores and it is relatively easy to do experiments like this.
Conducting market surveys is a third way to measure the demand for a product. Consumers are randomly selected and are asked questions related to how much of a product they’d buy at different