Bidding System Using Conversion Value and Rate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online systems optimize bids for content items based solely on conversion rates, overlooking the monetary value of products or services, leading to content items with high conversion values being less competitive due to low conversion rates.
Innovation Solution
An online system computes bid amounts based on the likelihood of user spend per conversion, incorporating minimum ROI and user-specific features through machine learning models that consider historical organic and attributed spending, allowing for optimization of monetary value over mere conversion numbers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If bids are optimized based on conversion rates, then the number of conversions is improved, but the monetary value of conversions deteriorates
Solution Approach 1:
The patent changes the bidding parameter from conversion rate alone to a composite parameter that includes both conversion rate and expected monetary value. The system calculates bid amounts by multiplying the expected monetary value by the conversion rate, thereby transforming the single-parameter optimization into a multi-parameter optimization that resolves the contradiction between conversion volume and conversion value.
2Reliability
If content items with high conversion values are targeted, then return on investment is improved, but competitiveness in auction deteriorates due to low conversion rates
Solution Approach 1:
The patent creates a composite bidding strategy that combines two previously separate metrics (monetary value and conversion rate) into a unified bid amount calculation. This composite approach allows content items with high monetary value to achieve competitive bid amounts even when their conversion rates are lower, thereby resolving the contradiction between ROI and auction competitiveness.
Data Source
AI summary
An online system calculates bids for content items to display to users based on the value of a product described in the content item and the likelihood of a viewing user purchasing the product. The online system identifies an impression opportunity for an ad request and computes an expected value of the conversion and a likelihood of the conversion. The online system computes a bid amount based on the expected conversion value and the likelihood of the conversion. Bids based on the value of the conversion allow a third party system offering the product to optimize for the value of each conversion instead of the conversion rate.