Machine Learning
Predictive Models
A predictive model is a mathematical model that makes it possible to predict future actions of clients or prospects. Predictive models are usually derived from historical customer data. For example, a model designed to identify customers who might terminate a subscription is typically based on the characteristics and past behavior of customers who have already terminated their contracts.
Why use a predictive model?
Predictive marketing models are mostly based on data mining techniques and the treatment of big data environments.
For example, here is a list of 7 types of predictive models:
- Acquisition models that predict the probability of converting a prospect into a customer
- Cross-selling models that predict the likelihood that an existing customer will buy an additional product or service
- The models of additional sale
- Attrition models
- The value models that are used to predict the different value measures such as the lifetime of a customer relationship or the value generated if the customer buys a specific product
- Tone patterns to predict what type of message will be best suited to each client
- Risk models that estimate the probability of fraudulent activity or credit problems
Additional Resources:
