Hashed Attribution Matching for Privacy-Preserving Ad Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User privacy concerns in mobile and web applications hinder effective data analysis for targeted advertising, as users increasingly opt out of sharing identifiable information, necessitating enhanced privacy technologies for accurate and anonymized attribution.
Innovation Solution
An advanced matching system separates impression and conversion datasets into buckets with minimum sizes, uses concatenated and hashed user identifiers, and computes intersections to maintain privacy while matching records.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If user identifiable information is collected for targeted advertising, then advertising effectiveness is improved, but user privacy is compromised
Solution Approach 1:
The patent extracts and removes personally identifiable information (PII) from advertising data through hashing and anonymization techniques. User identifiers are transformed into hashed values that cannot be reverse-engineered to original identities, thereby separating the useful advertising attribution function from privacy-sensitive information.
Solution Approach 2:
The patent introduces hashed user identifiers as an intermediary between raw user data and advertising analytics systems. This intermediary layer enables attribution tracking while preventing direct access to identifiable user information, acting as a privacy-preserving bridge between data utility and user anonymity.
2Object-affected harmful factors
If user data is anonymized to protect privacy, then user privacy is improved, but data matching accuracy deteriorates
Solution Approach 1:
The patent changes the parameter of user identifiers from plain text to hashed values through cryptographic transformations. This parameter change maintains the uniqueness and matchability of user identifiers while rendering them irreversibly anonymous, thus preserving both privacy protection and matching accuracy simultaneously.
Solution Approach 2:
The patent performs preliminary hashing and anonymization of user identifiers before they are used in advertising attribution processes. By pre-processing the data to remove identifiable information while preserving matching capabilities, the system ensures privacy is built-in from the start rather than added as an afterthought.
3Object-affected harmful factors
If multiple data buckets are created with minimum sizes for privacy, then user privacy is improved, but system complexity increases
Solution Approach 1:
The patent segments user data into different buckets or groups based on hashed identifier ranges or characteristics. This segmentation enables privacy-preserving analysis by aggregating data at appropriate levels while preventing identification of individual users, balancing privacy requirements with analytical utility through structured data organization.
Data Source
AI summary
Systems and methods herein described accurate and anonymized attribution. The described systems and methods access a set of impression data, access a set of conversion data, for every predefined time period, generates a salt value, generates a hashed set of impression data by appending the salt value to at least a portion of entries in the set of impression data, generates a hashed set of conversion data by appending the salt value to at least a portion of entries in the set of conversion data, discards the generated salt value, generates an intersection set of the hashed set of impression data and the hashed set of conversion data, determines a count based on the intersection set, and stores the count.