Location-Based Ad Selection Using Confidence Zones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current targeted advertising systems rely on keywords and location data, but they lack precision in delivering ads based on geographic accuracy, leading to inefficient ad placement and pricing models.
Innovation Solution
A system that receives geographic information from client devices, determines a potential location set, and selects ads based on the likelihood of the device being within a target geographic area, allowing for accurate ad delivery and dynamic pricing based on confidence values associated with location accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If advertisement selection is based on keyword matching only, then ad delivery is simple and fast, but geographic targeting precision is poor
Solution Approach 1:
The system segments the geographic target area into multiple zones with different confidence levels (e.g., high-confidence core area, medium-confidence surrounding area, low-confidence peripheral area). This segmentation allows the system to apply different pricing and selection strategies to different segments, improving geographic targeting precision without requiring complete redesign of the entire ad selection system.
Solution Approach 2:
The patent adds a new dimension to ad selection by incorporating geographic location and confidence levels alongside traditional keyword matching. This multi-dimensional approach (keyword + location + confidence) enables precise geographic targeting while maintaining the simplicity of keyword-based systems through layered decision-making.
2Measurement precision
If advertisement selection uses basic location data, then system operation is simple, but ad delivery accuracy to specific locations is insufficient
Solution Approach 1:
The system calculates confidence levels based on the overlap between potential device locations and geographic target areas, then uses this confidence feedback to dynamically adjust ad selection and pricing. This feedback mechanism transforms basic location data into actionable intelligence, improving delivery accuracy without complicating system operation.
Solution Approach 2:
The patent changes the parameter of location data from simple coordinates to a probabilistic confidence level that reflects the likelihood of the device being within the target area. This parameter transformation enables more accurate ad delivery by accounting for the uncertainty inherent in mobile device location data.
3Productivity
If all ads are priced uniformly, then pricing model is simple, but advertiser ROI optimization is limited
Solution Approach 1:
The system applies different pricing strategies to different geographic zones within the target area. High-confidence areas (where devices are most likely to be located) command higher prices, while lower-confidence areas use lower prices. This local quality approach optimizes advertiser ROI by matching price to delivery certainty without requiring a completely complex pricing model.
Solution Approach 2:
The patent implements a tiered pricing model where advertisers pay premium prices only for the portion of ad delivery that occurs in high-confidence geographic areas, rather than requiring uniform high pricing across all areas. This partial action approach allows advertisers to optimize ROI on high-probability deliveries while maintaining operational simplicity.
Data Source
AI summary
A system and method is provided that displays advertisements. In one aspect, advertisements are associated with target areas. When a request for information is received from a client device, the system and method determine a geographic area in which the client device may be located. Advertisements may be selected based on the likelihood that the client device is within the target area. In another aspect, advertisers may select prices for their advertisement that vary with such likelihood.