Dynamic Ad Selection System Using Performance Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Publishers face challenges in effectively managing and selecting advertisements for web pages, as existing systems lack the ability to express preferences over which ads to show, leading to inefficient ad targeting and performance optimization.
Innovation Solution
A process and system that allows publishers to define targeting criteria and selection parameters for content environments, generating request code to select and present advertisements based on explicit and implicit preferences, sequence, and performance metrics, enabling dynamic ad rotation and optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If context-based advertisement targeting is used, then advertisement selection is automated, but publishers cannot express preferences over which ads to show
Solution Approach 1:
The patent segments advertisement selection into multiple independent criteria: context-based targeting criteria, publisher preference criteria, and performance criteria. Each criterion can be independently defined and weighted, allowing the system to process diverse selection factors separately before combining them to select advertisements.
Solution Approach 2:
The patent implements dynamic advertisement selection by allowing publishers to define multiple advertisement groups with different priorities and by enabling real-time performance tracking. The system dynamically adjusts advertisement selection based on changing performance metrics and publisher preferences rather than using static context-based rules alone.
2Adaptability or versatility
If multiple advertisements are selected for an advertisement slot, then publisher preferences can be expressed, but the system complexity increases
Solution Approach 1:
The patent applies preliminary action by requiring publishers to pre-define advertisement groups, selection criteria, and performance thresholds before the advertisement selection process. This upfront configuration simplifies the real-time selection process, as the system only needs to evaluate predefined options against performance metrics rather than managing complex real-time negotiations between multiple criteria.
3Productivity
If advertisement selection is based on performance metrics, then click-through rates improve, but the ability to show diverse advertisements decreases
Solution Approach 1:
The patent changes parameters by introducing a multi-dimensional selection framework that considers not only performance metrics but also publisher preferences, context-based targeting, and advertisement group priorities. This multi-parameter approach allows the system to optimize for click-through rate while maintaining diversity through weighted criteria and rotational selection among multiple advertisement groups.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer program product, for selecting advertisements. In one aspect, a method includes receiving publisher selections of advertisements; associating the selected advertisements with an advertisement environment in a document; generating an advertisement request code for inclusion in the document; evaluating performance metrics for the selected advertisements; and in response to determining that the selected advertisements do not meet the performance threshold, optimizing the selection of selected advertisements based on the performance metrics; substituting a selected advertisement with a candidate advertisement and causing a client device to render the candidate advertisement in the advertisement environment in the document.