Customer-State Modeling
Description
Customer-State Modeling is a dynamic machine learning approach that predicts customer preferences/motives and the impact that the organizations marketing and customer initiatives have.
We may for example divide customers into different engagement-states based on past behavior and evaluate the probability (%) of customers being not-engaged and moving to an engaged state as a result of an organizations communication.
In sum, we predict how likely a customer is behaving in a certain way, even if we can't observe their current behavior directly, and act upon that.
Value
• Helps to evaluate the impact of marketing mix and customer experience interventions
• Helps to identify the customers that are more likely to engage with the organization
• Helps to identify less active customers that most likely can be prevented from ending the relationship with the organization
Applied in
• Non-contractual business models (B2C & B2B)
• Contractual business models (B2C & B2B)
Process
