Ad Selection via Customer Characteristic Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advertising methods for digital content, such as television programs, rely on predicted demographics and viewer numbers, leading to imprecise targeting and potential waste in advertising impressions, as they do not account for individual customer characteristics.
Innovation Solution
A system that selects advertisements based on historical customer data, including browsing and purchasing habits, to target specific customer characteristics, allowing for precise ad placement independent of the digital content requested, using a networked environment with a digital content server and advertisement selection application.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If advertising is sold based on predicted demographics and viewer numbers, then advertising coverage is broad, but advertising precision deteriorates
Solution Approach 1:
The patent segments the advertising system into distinct functional modules: a customer characteristics determination module that analyzes historical data, an advertisement selection module that matches ads to characteristics, and a presentation module that delivers content. This segmentation enables precise targeting by processing customer data through specialized components rather than using a monolithic predicted demographics approach.
Solution Approach 2:
The system performs preliminary determination of customer characteristics by analyzing historical browsing and purchasing data before the actual advertising presentation. This advance preparation creates a ready-to-use customer profile that enables immediate precise ad selection without requiring complex real-time analysis during content delivery.
2Productivity
If advertising is presented based on aggregate viewer predictions, then system operation is simple, but advertising effectiveness deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where customer characteristics are determined from historical data including past browsing and purchasing behavior. This feedback loop continuously refines the understanding of customer preferences, enabling increasingly effective ad targeting. The system learns from past customer actions to improve future advertising effectiveness without requiring proportional increases in system complexity.
3Measurement precision
If traditional demographic-based advertising is used, then data collection requirements are minimal, but advertising relevance deteriorates
Solution Approach 1:
The system enables customers to effectively 'self-profile' by analyzing their own historical browsing and purchasing data. The customer characteristics are determined automatically from their own past behaviors without requiring them to manually provide detailed information. This self-service approach generates highly relevant targeting data while minimizing the burden on customers and reducing the need for extensive active data collection.
Data Source
AI summary
Disclosed are various embodiments for selecting advertising for presentation to users in association with digital content items. Characteristics of a user are determined based at least in part on historical data that associates the user with one or more items. An advertisement is selected from an inventory of advertisements for presentation in association with a digital content item requested by the user. The advertisement is selected based at least in part on a highest value placed on the characteristics of the user by an advertiser.