Customer-Level Data Verification for Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing address verification systems often incorrectly match or mismatch customer data across multiple financial transaction instruments and accounts, leading to incorrect transaction risk calculations, declined transactions, and reduced customer satisfaction due to differences in billing addresses, mailing addresses, names, or telephone numbers.
Innovation Solution
A customer-level data verification system that compares data elements across multiple records to accurately verify customer information, reducing errors in transaction risk calculations and providing accurate comparison results for merchant decision-making processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional address verification systems compare customer data only at the account level, then the verification process is simple and fast, but verification accuracy deteriorates due to mismatches across multiple accounts
Solution Approach 1:
The patent segments the verification process into multiple levels: account-level verification and customer-level verification. The system divides customer records into multiple accounts, each with its own data elements, and performs verification at both the individual account level and the aggregated customer level. This segmentation allows the system to maintain simplicity for single-account customers while achieving high accuracy for multi-account customers by comparing data across all their accounts.
Solution Approach 2:
The patent merges data from multiple customer accounts into a unified customer-level view for verification purposes. By combining data elements across all accounts associated with a customer, the system creates a comprehensive verification baseline that accounts for variations in billing addresses, mailing addresses, names, and other data elements that naturally occur across different accounts. This merging resolves mismatches that would otherwise cause false rejections.
2Reliability
If the verification system strictly matches data elements, then fraud detection accuracy is high, but customer satisfaction deteriorates due to incorrectly declined transactions
Solution Approach 1:
The patent changes the verification parameter from strict exact-match requirements to flexible matching that allows for variations in data elements across multiple accounts. The system adjusts matching parameters to accommodate legitimate variations in billing addresses, mailing addresses, names, and other customer information while maintaining security by still detecting actual fraud patterns. This parameter adjustment reduces false positives that lead to incorrect transaction declines.
Solution Approach 2:
The patent introduces an intermediary verification layer that sits between the transaction request and the final authorization decision. This intermediary customer-level verification system mediates between strict fraud detection requirements and customer convenience by providing a more nuanced assessment that considers data variations across multiple accounts. It acts as a buffer that prevents outright rejections for legitimate variations while still blocking actual fraud.
3Productivity
If the system verifies data at the account level only, then processing speed is fast, but verification completeness deteriorates for customers with multiple accounts
Solution Approach 1:
The patent performs preliminary actions by pre-aggregating customer data across all accounts before the actual transaction verification occurs. The system maintains a pre-computed customer-level view that includes all data elements from all associated accounts, ready for rapid comparison during transaction processing. This preliminary preparation allows the system to provide comprehensive verification without adding significant processing time during the actual transaction, as the multi-account data is already organized and accessible.
Data Source
AI summary
A system, method, and computer program to reduce incorrectly declined transactions and improve risk calculation accuracy by reducing error probability during fraud detection. The tool first receives at least one data element as well as transaction account data and/or financial transaction instrument data. Then a customer is determined from a first record associated with the transaction account data and/or financial transaction instrument data. A record search is performed to identify at least one additional record associated with the customer. Finally, the data element is compared to the information contained in the additional record to create a comparison result that verifies a customer address. The comparison result may be used as an input to transaction risk calculations. The comparison result may also be provided to a merchant system and/or merchant for use in a decision-making process, for example, to verify customer identity.