Customer churn is the loss of clients or customers (also known as customer attrition, customer turnover, or customer defection). This concept was initially used within the telecommunications industry when many subscribers switched to other service providers. However, it has become a very important issue in other areas of business, such as banks, internet service providers, insurance companies, and so on. Well, two of the primary reasons for churn are customer dissatisfaction and cheaper and/or better offers from the competition.

As you can see in Figure 1, there are four possible contracts with the customer in a business industry: contractual, non-contractual, voluntary, and involuntary. The full cost of customer churn includes both the lost revenue and the (tele-) marketing costs involved with replacing those customers with new ones. However, this type of loss can cause a huge loss to a business. Think back to a decade ago, when Nokia was the dominator of the cell phone market. All of a sudden, Apple announced iPhone 3G, and that was a revolution in the smartphone era. Then, around 10 to 12% of customers stopped using Nokia and switched to iPhone. Although later on, Nokia also tried to release a smartphone, eventually, they could not compete with Apple:

Churn prediction is fundamental to businesses, as it allows them to detect customers who are likely to cancel a subscription, product, or service. It can also minimize customer defection. It does so by predicting which customers are likely to cancel a subscription to a service. Then, the respective business can have a special offer or plan for those customers (who might cancel the subscription). This way, a business can reduce the churn ratio. This should be a key business goal of every online business.

When it comes to employee churn prediction, the typical task is to determine what factors predict an employee leaving his/her job. These types of prediction processes are heavily data-driven and are often required to utilize advanced ML techniques. In this chapter, however, we will mainly focus on customer churn prediction and analysis. For this, a number of factors should be analyzed in order to understand the customer's behavior, including but not limited to:

• Customer's demographic data, such as age, marital status, and so on
• Customer's sentiment analysis of social media
• Browsing behavior from clickstream logs
• Historical data that shows patterns of behavior that suggest churn
• Customer's usage patterns and geographical usage trends
• Calling-circle data and support call center statistics