Modeling Price Effects Using Point of Sales Data

While online experiments are an effective way to test the effect of price on demand in a realistic setting, they can accommodate only a limited set of factors and their respective levels. In reality, demand is a function of not only price but also other factors such as product quality and features, marketing communication (e.g., messages and intensity), and the marketing channels (e.g., number of territories, stores, salesforce). Therefore, an expanded view of a demand function is one that relates demand to not only price but also other marketing mix variables as well as competitive information. Having such an expression of demand function helps managers in evaluating the effectiveness of total marketing efforts, not just price. A comprehensive demand function also helps the firm predict demand with greater certainty, and plan for its sourcing, production, inventory, and salesforce allocations. Finally, such a demand function helps the firm with optimal resource allocation decisions. For instance, the firm can investigate whether a price reduction or an increase in advertising results in greater revenue and profits. With increased availability of actual sales or consumption data at the point of sales (or consumption), analysts are able to use realistic data to investigate the effects of price on demand.