Measuring Predictive Power

This chapter explains common methods in evaluating model predictive power. If the goal is defined as finding the most important/risky customers, there are many different ways using the available resources. Analysts measure accuracy and look for answers. It is obvious that two different analysts would provide different models; however, what both are looking for is an adequate level of accuracy. That means that analysts have freedom while looking for models, but the final model needs to be accurate and usable for decision making. No matter what the final model is, the most important factors before the final results are confirmed are the model relevance tests. One can, for example, create several models with the same goal but using different methods or methodologies. The one with highest accuracy level is the best one. It is important to point out that models do not have to be based only on one method but can combine several methods at the same time.

Social Network Analysis (SNA) for Churn Mitigation/Detection

Social network analysis is intentionally covered in a separate chapter for two reasons. First, the importance of this method has rapidly increased in past few years, and second, there are very few useable studies that cover social network analysis concepts in churn management. By understanding the methods explained in Chapter 3 and combining them with knowledge of SNA concepts, the analysts (readers) can unlock the full potential of advanced analytics in one of the most important fields of research today, customer relationship and especially churn analysis. With the ability to understand how those metrics can be used, integration of those methods into more complex environments is explained regarding the key topic, churn management.

Setting (Realistic) Business Aims

This chapter explains different perspective of churn analysis and points out the importance of understanding what really can or cannot be done. In addition, it is important to understand common errors analysts (readers) have, so that one can be aware of them when planning and conducting churn analyses. It is advisable for the reader to move back to the introduction and Chapter 1 after finishing reading in order to once again understand the full potential and restrictions of the proposed methods and techniques. Although this chapter covers churn topics on a conceptual level, it is very important for the reader to be able to understand and express key points on this level. By using industry-related cases and by combining churn with early warning systems, the complete scope is covered, and the reader can move to the next level, techniques, explained in next chapter.

Churn Analysis Using Selected Structured Analytic Techniques

After explaining how to prepare data, an introduction to structured analytic techniques is covered in this chapter. The importance of structured techniques comes from their simplicity and wide usage, making them fast to use and efficient to structure in even complex environments. For further explanation of how those techniques could be applied in the business environment, analysts (readers) should look closer into the case studies in last chapter of this book. Besides the ability to structure logic problems, analysts (readers) also needs to be aware of the different motivators influencing market conditions, especially from the customers' perspective. The chapter ends with a brief introduction of consumer behavior, making it part of the churn topic.

Data Preparation and Churn Detection

This chapter describes data preparation techniques for different churn models. The central topic is data sampling as preparation for building churn models, especially for predictive models. The chapter shows how to construct a data sample that will reflect business reality and show good performance regarding building predictive models. A significant part of the chapter is dedicated to construction of derived variables, which are a direct reflection of expert knowledge used within churn models. Beside data preparation for predictive models, the chapter also describes data preparation techniques for other methods usable for churn modeling like survival models, fuzzy expert systems, K-mean clustering, etc. The attribute relevance analysis chapter described different techniques for attribute importance detection usable in churn modeling. It gave descriptions with examples of how to make an attribute relevance analysis for predictive churn models in case of binomial target variables, as well in case of multinomial target variables. This chapter covers dummy variable construction and profiling techniques based on attribute relevance analysis, as well as logic checks from the perspective of business users.

Churn Model Development, Monitoring, and Adjustment

This chapter is based on the fact that the finalization of the model building stage is the beginning of the periodic monitoring and redesigning stage. The churn solution should be adopted by market changes, internal company policy changes, portfolio structure changes, and other factors. The chapter gives answers about monitoring frequency and techniques with which the company could realize when to change into the existing churn solution. Another important topic covered in this chapter is “what if” analysis techniques, how to make scenarios for future churn trends regarding planned changes while taking in consideration the current state of the existing portfolio. The chapter ends with business strategy creation based on revealed knowledge from the churn solution and explains the importance of cooperation between business sectors and analysts in all stages of churn solution development from planning and realization to usage.

Attribute Relevance Analysis

Structured problems need to be quantified by relevance, and that is explained in this chapter. The most common methods of relevance analysis and different strength and sensitivity measures are explained in a way that practitioners can easily use them and start experimenting. Contrary to the belief that powerful hardware and sophisticated software can substitute for attribute relevance analysis, attribute relevance analysis is an important part of each analysis that operates with the target variable. Recognition of the most important variables, those with the greatest impact on the target variable, reduces redundancy and uncertainty at the model development process stage. It provides robustness of the model and model reliability. Attribute relevance analysis also evaluates attribute characteristics. Attribute characteristics evaluation includes measuring attribute values' impact on target variables. It helps in understanding relations and logic between the most important predictors and the target variable and understanding relations and logic between the most important predictors from the target variable perspective. Making models relevant and being able to proof them is relevant and almost as important as the ability to build them. After completing this chapter, analysts (readers) are ready to start projects.

From Churn Models to Churn Solution

This chapter explains churn model classification, describes techniques for developing predictive churn models, and describes how to build churn segmentation models, churn time-dependent models, and expert models for churn reduction. Analysts (readers) are shown a holistic picture for churn modeling and presented an analytical method with techniques described as elements that could be used for building a final churn solution depending on current business problems and expected outputs. There are numerous ways for designing final churn models (solutions). The first criteria is to find solutions that will be in line with business needs. The problem is not applying some data mining technique; the problem is in choosing and preparing appropriate data sets. Applied techniques should show holistic solution pictures for churn, which are explainable and understandable for making decisions, which will help in churn understanding and churn mitigation.

Churn Problem in Everyday Business

This chapter is an introduction into customer relationship management, explaining the modern business environment and techniques to monitor it. As part of that process, churn management is introduced and explained across industries. Throughout the chapter, a wider churn perspective is explained together with several examples from real cases. As the chapter comes to its end, the business approach become more and more involved, and the reader starts to realize the importance of churn management and its complexity. At the same time, the idea of techniques and methodologies as to how it can be managed start to shape key book points.

Churn Case Studies

This case study chapter brings two business cases in the domain of churn, both unique in many ways, combining almost all the topics covered inside book. The first business case presents a retail company facing new competitors and consequently preparing a customer-retention strategy. The case introduces the business environment in which the company was operating prior to the arrival of new competitors while the model is being devised for the purpose of preventing or at least buffering the churn trend as a reaction to the new competition. Development of an early warning indicator system based on data mining methods is also described as a support to the management in the early detection of both market opportunities and threats. The second business case describes the situation in a telecommunication company in the domain of churn prediction and churn mitigation. The churn project was divided into a few stages and is fully described in the chapter. The case explains how the company can decrease the churn rate and gives directions for better understanding of customer needs and behaviors.