Real-Time Bidding System for Cost Per Engagement Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Real-Time Bidding environments for online advertising, determining which Media Properties to bid on, how frequently to bid, and what prices to bid can be challenging to achieve optimal cost per engagement and campaign success, requiring strategies for pre-campaign optimization and dynamic re-evaluation during campaign runtime.
Innovation Solution
Systems and methods for optimizing online advertising campaigns by stack ranking Media Properties based on cost-per-engagement, assigning impact levels, and dynamically adjusting bidding strategies based on engagement rates and campaign progress to allocate budgets effectively across diverse display formats and Media Properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If bidding strategies are manually determined for each Media Property, then bidding decisions can be customized, but the complexity of managing billions of auction opportunities increases significantly
Solution Approach 1:
The patent segments the vast set of Media Properties into distinct groups or categories, allowing the system to manage billions of auction opportunities by treating similar properties uniformly rather than individually. This segmentation enables customized bidding strategies to be applied to groups rather than each individual property, reducing system complexity while maintaining precision.
Solution Approach 2:
The system changes bidding parameters dynamically based on campaign performance data. By adjusting bid amounts, frequency, and targeting parameters automatically based on measured engagement metrics, the system achieves precise bidding control without requiring manual configuration for each Media Property, thus resolving the complexity issue.
2Productivity
If bidding frequency is increased to capture more opportunities, then campaign fulfillment improves, but cost per engagement increases
Solution Approach 1:
The system dynamically adjusts bidding frequency based on real-time campaign performance and Media Property characteristics. Rather than using a fixed high frequency, the system optimizes bid timing and frequency automatically, increasing bids when engagement likelihood is high and reducing them when cost efficiency deteriorates, thus achieving both fulfillment and cost control.
Solution Approach 2:
The system implements feedback loops that monitor engagement outcomes and use this information to adjust future bidding behavior. By analyzing which bids resulted in engagements and which did not, the system learns to optimize bidding frequency and timing, reducing wasted spend on low-probability opportunities while maintaining adequate fulfillment rates.
3Reliability
If bid prices are increased to win more auctions, then impression acquisition improves, but cost per engagement deteriorates
Solution Approach 1:
The system applies differentiated bid pricing strategies to different Media Properties and auction contexts rather than using uniform high bids. By analyzing local characteristics of each opportunity (Media Property quality, audience engagement history, campaign objectives), the system determines optimal bid levels locally, ensuring reliable acquisition for high-value opportunities while avoiding overpayment for lower-value ones.
Solution Approach 2:
The system employs partial bidding strategies where it may bid aggressively on a subset of high-priority opportunities while bidding conservatively or skipping lower-priority ones. This selective approach ensures adequate impression acquisition for the most valuable opportunities without incurring excessive costs across all auctions, optimizing the balance between reliability and cost efficiency.
Data Source
AI summary
Systems and methods are disclosed for optimizing an online advertising campaign both before the campaign begins, and dynamically during the campaign. Optimizations are performed comparatively between a plurality of MPs (Media Properties) based on their relative cost-per-engagement. Comparisons are performed by first stack ranking MP inventory including any of sites, feeds, and verticals, based on cost per engagement. Once ranked, scores are assigned to the targeted inventory and a mean score is determined. Then, the inventory is rated as high, normal, or low impact based on their scores compared with the mean and a standard deviation for all scores. Higher impact sites with scores at least a standard deviation above the mean are initially favored, and the MP targeting strategy is dynamically adjusted during the campaign based on periodically re-evaluating the MP rankings, frequencies of engagement, and campaign progress relative to fulfillment in an allotted run time.