Chargeback Dispute Scoring Model for Payment Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack the ability to provide a score indicating the probability of the outcome of a chargeback dispute, forcing issuers and merchants to inefficiently decide on whether to submit or dispute chargebacks, leading to resource wastage and increased network traffic.
Innovation Solution
A chargeback dispute scoring system that generates a prediction model using historical transaction and account profile data to determine a chargeback dispute score, indicating the likelihood of a favorable outcome for merchants or issuers, thereby informing decision-making and reducing unnecessary disputes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If issuers and merchants submit or dispute all chargebacks without scoring, then they ensure thorough review of all transactions, but this leads to resource wastage and increased processing time
Solution Approach 1:
The system performs preliminary analysis by generating chargeback dispute scores before actual disputes are filed. The scoring model evaluates historical transaction data, account profile data, and chargeback characteristics in advance to predict dispute outcomes, allowing issuers and merchants to make informed decisions about whether to proceed with disputes, thereby avoiding unnecessary processing time and resource expenditure
2Loss of energy
If issuers and merchants dispute all chargebacks, then they maximize their chances of recovering funds, but this increases network traffic and processing resources required
Solution Approach 1:
The system implements feedback by providing chargeback dispute scores to issuers and merchants based on historical data analysis. This feedback mechanism enables parties to adjust their dispute strategies by understanding the predicted outcome probabilities, allowing them to allocate banking resources more efficiently by focusing disputes on cases with higher success probabilities rather than disputing all chargebacks uniformly
3Measurement precision
If a prediction model is generated using historical data, then decision-making accuracy is improved, but system complexity increases
Solution Approach 1:
The scoring system achieves universality by creating a multi-functional prediction model that processes multiple types of input data (transaction data, account profile data, chargeback characteristics) and generates comprehensive dispute scores. This single system serves multiple purposes: predicting dispute outcomes, guiding dispute decisions, and optimizing resource allocation across different issuers and merchants, thereby justifying the system complexity through its broad applicability and enhanced measurement precision
Data Source
AI summary
A dispute scoring computing device for generating chargeback dispute scores is provided. The dispute scoring computing device extracts historical transaction data and account profile data for a plurality of transactions associated with an account, and generates a chargeback dispute prediction model for the account based on the extracted data. The dispute scoring computing device further receives a candidate chargeback for a transaction initiated with a merchant using the account, extracts transaction data from the candidate chargeback, and determines, using the generated chargeback dispute prediction model, a chargeback dispute score indicating a likelihood that the candidate chargeback, if disputed by the merchant, would result favorably for the merchant.