Revenue Sharing Bid Valuation via Cookie Attribution
Solution Overview
Problem
Current online advertising systems face challenges in efficiently managing revenue-sharing bids, particularly in valuing and pricing both fixed and revenue-sharing bids in auctions, and accurately attributing user actions to advertisements to ensure fair charging of content providers.
Innovation Solution
The system receives bids for content slots, determines winning bids, logs presentation details, and charges content providers based on user actions, using cookies to track selections and actions, allowing for accurate revenue allocation and fair pricing in mixed auctions combining cost-per-click and revenue-sharing models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If revenue-sharing bids are used to price content providers, then fair charging based on actual user actions is improved, but the complexity of valuing and pricing bids in auctions increases
Solution Approach 1:
The advertising system acts as an intermediary between content providers and users, tracking user actions through cookies and redirect URLs to accurately attribute conversions to specific advertisements. This mediator enables precise revenue sharing calculation without requiring direct complex tracking between all parties.
Solution Approach 2:
The system implements feedback loops where user actions are tracked, attributed to specific advertisements through cookies and redirect URLs, and used to calculate revenue sharing payments. This feedback mechanism enables accurate charging based on actual performance while maintaining manageable system complexity through automated processing.
2Measurement precision
If cookies are used to track user selections and actions, then accurate attribution of user actions to advertisements is improved, but the complexity of tracking and managing session data increases
Solution Approach 1:
The system uses cookies as digital copies that store user session information and tracking data on users' devices. These cookie copies enable the advertising system to track and attribute user actions without requiring complex real-time communication channels or storing all data centrally, thus reducing tracking complexity while maintaining attribution accuracy.
Solution Approach 2:
The system performs preliminary tracking actions by setting cookies during initial user sessions, establishing tracking identifiers before user actions occur. This preliminary setup simplifies subsequent attribution processing by having tracking data already in place when user actions need to be measured.
3Adaptability or versatility
If mixed auctions combining cost-per-click and revenue-sharing models are used, then versatility of bidding options is improved, but the difficulty of valuing and pricing bids in auctions increases
Solution Approach 1:
The auction system segments bid processing by separating cost-per-click bids from revenue-sharing bids, handling each type through dedicated valuation logic. This segmentation allows the system to maintain versatility in accepting multiple bidding models while reducing the complexity of bid valuation by processing each model type through standardized, isolated procedures.
Solution Approach 2:
The system changes valuation parameters dynamically based on bid type, applying cost-per-click pricing logic to CPC bids and revenue-sharing calculation logic to revenue-sharing bids. This parameter adaptation enables the system to handle diverse bid models uniformly, reducing the perceived complexity by treating different bid types through standardized parameter transformation rather than separate processing paths.
Data Source
AI summary
A bid for a content slot from a content provider is received. The bid includes a revenue-sharing bid. It is determined that the bid is a winning bid and content is presented from the content provider in the content slot. Presentation details are logged that include a time when the content was presented. At a later time, an indication is received of an action, the indication including an indication that the action is responsive to the content having been presented previously. Based on the action, the content provider is charged an appropriate amount based on the revenue sharing bid.