CORTEX aiCHURN is an end-to-end integrated AI-based solution that provides churn prediction and mitigation. Churn, which is the loss of customers to competition, is a problem for companies across multiple industries, because it is more expensive to acquire a new customer than to keep your existing one from leaving. Most consumer-facing companies suffer from voluntary churn. The churn rate has strong impact on the life time value of the customer because it affects the length of service and the future revenue of the company. For example, if a company has 20% churn rate, then the average customer lifetime is 5 years. Similarly, a company with a churn rate of 50%, has an average customer lifetime of 2 years. Churn effectively erodes a company’s profitability. A company typically spend significant money to acquire a new customer and when that customer leaves, the company not only loses the future revenue from that customer but also the resources spend to acquire that customer.
The CORTEX aiCHURN solution focus on churn reduction via state-of-the-art AI and can predict a customer's propensity to churn by using information about the customer such as household and financial data, transactional data, and behavioral data. CORTEX aiCHURN also addresses a related problem of lapse in payment or non-activity which can lead to churn or non-active customers. So models that predict a customer’s lapse propensity or non-activity could be used in proactive mitigation strategies.
In one of the use cases, a leading insurance brand in the “call center” market experienced a 40% cancellation rate of all their policies issued within 90 days after the sale. A churn prevention solution was developed using a variety of customer data sources such email correspondence, voice recordings, text interactions and demographics. The “proposed sale” was then analyzed and the price adjusted to reflect the risk and or reject the sale the next day before the bulk of the costs relating to the issue of a policy is incurred. This lead to significant cost savings for the insurer.
See Figure 15 for another use case where CORTEX aiCHURN was implemented to predict lapse propensity and non-activity on a funeral policy insurer. The data sets consisted of funeral policy data for main members and dependents as well as the payment transactions. Apart from accurate predictive models that generalized well on unseen test data, insightful analysis was also provided such as the average probability to lapse by age, gender, plan brand, plan premium, etc.
In a telecommunication use case, the company had a problem of annual churn rates of about 10%, while the customer base grew at a slower rate of 5% annually. Some of the factors impacting churn included customers with a significant number of service calls, those with higher bills and those with international plans. A churn prediction model helped to capture a significant portion of the potential churners, which lead to proactive actions to reduce the churn rate by up to 50%, leading to significantly revenue increases.
In a mobile money use case, a reduction of churn and improved customer loyalty was ensured via the use of predictive churn models and increasing mobile wallet activity with attractive personalized product offerings.
Relevant industries : Financial services, Healthcare, Retail & eCommerce, Telecommunications, Media & Entertainment, Education, and Travel & Tourism.