Dual Blind Attribution System for Privacy-Safe ROI Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring the impact of online advertising on offline sales require sharing consumers' personally identifiable information, which is privacy-invasive and legally and security-wise undesirable, necessitating a solution that maintains consumer privacy while still enabling effective ROI analysis for vendors.
Innovation Solution
A dual-blind method and system where a browser identifier and temporary ID are used to attribute user activity without revealing personal information, allowing analytics servers to match user profiles without tracing back to the original consumer, ensuring privacy while enabling ROI analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If personal identification information is shared among multiple parties to enable ROI analysis, then measurement precision is improved, but consumer privacy is compromised
Solution Approach 1:
The patent introduces multiple intermediary components including a content delivery network (CDN) that issues opaque identifiers, an attribute server that stores consumer attributes without linking them to identifiers, and an analytics server that performs matching without accessing personal information. These intermediaries enable precise attribution while maintaining privacy by design, as each component handles only non-personally-identifiable data except the retailer which directly collects PII for transaction purposes but not for attribution analysis
Solution Approach 2:
The system segments the attribution process into distinct functional modules: (1) identifier generation by CDN, (2) attribute storage by attribute server, (3) profile retrieval by user profile server, and (4) matching by analytics server. This segmentation allows each module to operate with minimal data exposure, where personal information is isolated to the retailer's direct interaction with consumers and never shared with analytics components, thus resolving the contradiction between measurement precision and privacy protection
2Productivity
If personal identification information is captured and stored by vendors, then productivity is improved through better marketing data, but security requirements and legal compliance complexity increase
Solution Approach 1:
The patent extracts personally identifiable information from the attribution workflow entirely. The system uses opaque identifiers issued by the CDN that cannot be traced back to consumers, and stores consumer attributes (demographics, preferences) in an attribute server without linking them to these identifiers. This extraction eliminates the need for vendors to capture and store sensitive PII while maintaining the ability to perform sophisticated marketing analysis through attribute-based segmentation
Solution Approach 2:
The system employs temporary, disposable opaque identifiers that are issued by the CDN and used solely for attribution purposes. These identifiers have no intrinsic meaning and cannot be used to reconstruct consumer identity, effectively serving as single-use tokens that enable tracking without creating long-term privacy risks or compliance burdens for vendors
Data Source
AI summary
A method and system for attributing activity to a user includes sharing information with an analytics server while attributes and a user profile cannot be tracked back to the original user from the analytics server. A temporary ID included in the shared information is destroyed, thus eliminating any trace back.