Behavioral Lift Calculation for Ad Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online digital advertising systems fail to effectively select relevant advertisements for users, as they primarily rely on search queries without considering users' historical behavior patterns, leading to inefficient ad targeting.
Innovation Solution
An advertising selection system that develops and maintains user profiles based on various behaviors, including webpage content, search queries, purchases, and advertiser interactions, to calculate the 'lift' of specific behaviors and build models for consumer conversions, enabling personalized ad targeting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If advertising systems rely only on search queries to select advertisements, then the system complexity is low, but the advertising relevance and conversion effectiveness deteriorate
Solution Approach 1:
The patent segments user behavior into multiple dimensions including search queries, webpage content, mobile application usage, purchases, transactions, and advertiser website activities. Each dimension is analyzed separately to calculate behavior-specific conversion rates, which are then integrated to form a comprehensive user profile for more accurate ad selection.
Solution Approach 2:
The system performs preliminary actions by maintaining and updating user profiles with historical behavior information before ad selection occurs. Conversion rates for different behaviors are pre-calculated and stored, allowing the system to quickly retrieve and apply relevant behavioral data when selecting advertisements, rather than computing everything in real-time.
2Measurement precision
If the system calculates lift for all behavior tuples iteratively, then the behavioral model accuracy improves, but the computational time and resources increase
Solution Approach 1:
The patent applies partial action by calculating lift values for behavior tuples in a systematic iterative process, where the full computational burden is distributed across multiple processing cycles. The system calculates lift for individual behaviors first, then progressively evaluates behavior tuples, stopping when the marginal gain in model accuracy no longer justifies the computational cost.
Solution Approach 2:
The system dynamically adjusts its computational approach by recalculating lift values for remaining behaviors after each behavior tuple is identified and removed from the population. This dynamic recalculation continues adaptively, allowing the system to achieve high accuracy for the most significant behavioral patterns while avoiding unnecessary computation for less impactful behaviors.
3Measurement precision
If user profiles are updated with comprehensive behavior information from multiple sources, then the ad targeting precision improves, but the data processing complexity increases
Solution Approach 1:
The patent creates a universal user profile structure that can accommodate multiple types of behavior data from different sources including search queries, webpage content, mobile applications, purchases, transactions, and advertiser website activities. This unified profile format allows the system to process diverse data types through a common framework, reducing the complexity that would otherwise arise from handling each data source separately.
Solution Approach 2:
The system incorporates feedback mechanisms where conversion information from advertisers is used to update and refine user profiles. The calculated lift values and behavioral models provide feedback that continuously improves the accuracy of ad targeting by adjusting the weight and significance of different behavior types based on their actual conversion performance.
Data Source
AI summary
An advertising system identifies behaviors from user activity and associates the behaviors with a user profile. Advertisers provide the advertising system with information on conversion rates of users associated with user profiles. A behavioral model of user responses is built to identify the relative frequency of behaviors for increasing the response rate of ads. Incoming advertising requests are matched to modeled behaviors to determine an advertiser's interest in bidding on the ad placement.