MAJOR CHALLENGES OF PRODUCT-BASED SEGMENTATION

An advantage of customer-based segmentation versus product-based segmentation is that it is typically easier to identify which consumers belong to each segment. Simply checking the birthdate on a person’s driver’s license, for example, can confirm whether or not the person is eligible for a senior discount. As previously discussed, however, product-based segmentation has a lot of advantages over customer-based segmentation, and is typically worth the additional effort required to identify the micro-segments and build versions of the product specifically designed for each segment.

THE DAWN OF BIG DATA AND BUSINESS ANALYTICS

The majority of the segmentation categories and methods discussed so far have been around for at least the last 20 years. The science in this area has been re-energized recently, however, with the emergence of Big Data and Business Analytics. Big Data describes the vast amount of unstructured data depicting customer preferences that has just recently become available from social network sources such as Facebook, Twitter, and, from customer web search history. The availability of this type of data combined with an exponential increase in computing power provides new opportunities for categorizing customers into more refined segments. In addition, the econometric and data mining tools have also had significant advances in recent years. The science of applying these tools to the new opportunities provided by Big Data is commonly termed Business Analytics. Business Analytics can be characterized into three different categories:

  • 1. Descriptive Analytics.
  • 2. Predictive Analytics.
  • 3. Prescriptive Analytics.

Descriptive analytics involves the science of identifying different customer segments such as the ones described in this chapter. While this practice remains challenging, it is the area where the most advances have already been made and represents the most prevalent uses of business analytics in practice. A simple way of thinking about descriptive analytics is as a way of better understanding who your customers are. It allows firms to think of their customer base as a combination of many micro-segments, so as to design targeted products and advertising programs for each segment.

Predictive analytics is closely related to descriptive analytics except for the main objective, which is to predict customer demand or their reactions to a set of marketing exposures. Time series forecasting, where past demand data are extrapolated into the future using statistical techniques, is a subset of predictive analytics. Time series forecasting is discussed in Chapter 2. Prescriptive analytics goes beyond time series forecasting, however, to include causal variables such as price, promotions, weather, economic conditions, and other possible predictive variables. Both predictive and prescriptive analytics often rely on econometric techniques such as regression analysis, so it is sometimes confusing to distinguish the two. A simple way to differentiate between the two is the following. If your end goal is simply to make better predictions without a need to understand the causal reasons for the outcomes, then you are employing prescriptive analytics. On the other hand, if you are more concerned with understanding the underlying causes of some outcome (such as sales), then you are employing prescriptive analytics. From a slightly more technical standpoint, predictive analytics involves studying the significance of the possible explanatory variables in a regression model while predictive analytics is only concerned about the predictive accuracy of the model. Both predictive and prescriptive analytics methods, as they relate to pricing and promotion, are discussed are discussed in Chapters 3, 7 and 8.

Prescriptive analytics describes the science of using the forecast provided by predictive analytics in an optimization model to guide firms on how to set prices or allocate capacity so as to achieve some objective such as maximizing profits or market share. In capacity-based revenue management (described in Chapter 4), the objective is to save expiring capacity such as an airline seat for latter arriving segments who are willing to pay more for the product. In pricing analytics (described in Chapters 6, 7 and 8), the objective is to set prices or target promotions so as to maximize the firm’s overall profits. Prescriptive analytics, as with most business tools, starts with a forecast—the topic of the next chapter.

 
Source
< Prev   CONTENTS   Source   Next >