Self-supervised churn prediction using profitability clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Financial institutions face challenges in accurately identifying customer churn due to customers' ability to switch banks without notice, and existing methods often target the wrong customers with marketing campaigns, resulting in wasted resources.
Innovation Solution
A system and method for self-supervised churn predictions using two machine learning models: the first model clusters customers based on profitability features over historical time intervals, and the second model predicts future clusters, flagging customers likely to churn by moving to lower profitability clusters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If deterministic rules based on domain knowledge are used to identify customer churn, then marketing campaigns can be targeted, but the accuracy of churn identification deteriorates and resources are wasted
Solution Approach 1:
The patent replaces deterministic mechanical rules with machine learning models that automatically learn patterns from data. The first model clusters customers based on profitability features, and the second model predicts future cluster assignments, enabling accurate churn prediction without manual rule-setting.
Solution Approach 2:
The patent transforms static domain knowledge into dynamic machine learning models that continuously adapt to changing customer behaviors. The models process multiple features including transaction patterns, account balances, and temporal patterns to dynamically identify churn risks.
2Adaptability or versatility
If customers are allowed to switch banks without notice, then customer freedom is improved, but the ability to track customer behavior and identify churn deteriorates
Solution Approach 1:
The patent performs preliminary actions by continuously monitoring and clustering customers based on their behavior patterns before churn occurs. The system identifies customers at risk of churning by analyzing changes in their transaction patterns, account usage, and profitability metrics, allowing proactive intervention before customers actually leave.
Solution Approach 2:
The system establishes feedback loops where customer behavior data continuously feeds back into the machine learning models to refine churn predictions. The models learn from historical data and update their predictions in real-time, creating a dynamic system that adapts to changing customer behaviors without requiring explicit churn notifications.
3Measurement precision
If monthly subscription fees or affirmative notification requirements are imposed, then churn identification accuracy is improved, but customer convenience and business attractiveness deteriorate
Solution Approach 1:
The patent implements self-service by having the system automatically monitor, cluster, and predict churn for customers without requiring any action from them. The machine learning models autonomously process customer data, identify patterns, and flag potential churn cases, eliminating the need for customers to manually notify the bank or pay subscription fees for monitoring services.
Data Source
AI summary
Features from historical transaction and customer data of a financial institution are calculated and/or extracted. Each customer is assigned to a given profitability cluster within each interval of time over a historical period of time based on the corresponding features. A self-supervised machine learning model is trained on the features to predict the clusters in a future interval of time. Features for a most-recent past interval of time are provided as input to the model and the model returns a predicted cluster for a given customer in a future interval of time. When the customer-assigned cluster in the most-recent past interval of time is a higher prioritized cluster than the predicted cluster for the future interval of time, a system of a financial institution (FI) is notified to take one or more mitigating in an attempt to prevent customer churn with the FI.