Bid Amount Modification for Conversion Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online systems struggle to reliably determine the likelihood of users performing conversions associated with sponsored content, leading to inefficient targeting of advertisements to users who are likely to perform specific actions.
Innovation Solution
The online system modifies the bid amount for advertisement requests based on a subsidy value, which increases the likelihood of presenting ads to users who are likely to perform the desired action, while a penalty value offsets the subsidy across multiple opportunities, ensuring efficient budget allocation and preventing price exceeding the bid amount.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the online system uses conventional bid amounts based on ad access only, then the selection process is simple, but the system cannot reliably target users likely to perform conversions
Solution Approach 1:
The system performs preliminary actions by determining user likelihood scores for conversions before the bid amount is finalized. Conversion likelihood data is pre-calculated based on user profiles, historical behavior, and predicted actions, allowing the bid modification to incorporate these insights without adding complexity to the real-time selection process
Solution Approach 2:
The system introduces an intermediary mechanism - the bid modification value - that bridges the gap between conventional bid amounts and conversion likelihood. This intermediary translates complex user behavior analysis into a simple multiplicative factor that adjusts the bid amount, maintaining system simplicity while improving targeting accuracy
2Reliability
If the online system increases bid amounts for users likely to convert, then conversion targeting improves, but the price may exceed the original bid amount
Solution Approach 1:
The system implements feedback by continuously monitoring the relationship between modified bid amounts and actual conversion outcomes. The bid modification value is adjusted based on observed conversion rates, ensuring that increased bids for high-likelihood users actually deliver value and preventing systematic overspending
Solution Approach 2:
The system dynamically changes the bid amount parameter based on user-specific conversion likelihood. Rather than using a fixed bid, the bid is transformed into a variable parameter that scales with predicted user value, allowing the system to pay more for high-value interactions while maintaining budget control through the original bid as an anchor
3Productivity
If the online system determines bid amounts based on predicted user actions, then ad selection optimizes for conversions, but the system cannot access reliable conversion likelihood data
Solution Approach 1:
The system prepares for data limitations by establishing default bid modification values and fallback mechanisms. When conversion likelihood data is unavailable or unreliable, the system cushions against this information gap by using conservative estimates or default behaviors, ensuring the bid modification process can continue without failing due to missing data
Data Source
AI summary
An online system receives a sponsored content item including a maximum amount of compensation for accessing the content, a budget, and a tracking mechanism identifying an action. When an opportunity to present sponsored content to a user eligible to be presented with the sponsored content item is identified, the online system determines a likelihood of the user performing the action identified by the tracking mechanism and an average likelihood of other users performing the action identified by the tracking mechanism. Based on the determined likelihood and the average likelihood, the online system determines a subsidy value. Additionally, the online system generates a penalty value inversely proportional to a number of the identified action that have been identified. The online system increases a bid amount by the subsidy value decreases the bid amount by the penalty value to determine whether to present the sponsored content item to the user.