User Activity-Based Ad Targeting via Bid Auction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional web advertising methods lead to oversubscription of significant pages and undersubscription of less significant pages, resulting in inefficient ad revenue and low advertising effectiveness, as they do not accurately target user interests based on their activities.
Innovation Solution
A software facility that identifies distinct audiences based on user activities on a publisher's website and allows advertisers to bid on opportunities to display targeted ads to these audiences across any page, using auction systems or fixed pricing to ensure better ad placement and revenue generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If advertisers target specific significant pages on a publisher's website, then advertising revenue from those pages is maximized, but significant pages become oversubscribed and less significant pages remain undersubscribed
Solution Approach 1:
The patent segments the advertising opportunity from specific web pages and reassigns it to user-based audiences. Instead of tying ads to particular pages (significant or not), the system divides the website's audience into segments based on their activities and interests, allowing ads to follow users across any page they visit. This resolves the contradiction by making advertising placement flexible (principle of adaptability) while maintaining revenue efficiency through precise user targeting.
Solution Approach 2:
The patent transitions from a two-dimensional page-based advertising model to a three-dimensional user-across-pages model. By adding the temporal dimension of user navigation and the dimensional aspect of user interests/activities, the system allows advertising to occur in any location on the website where the target user appears, rather than being constrained to predetermined significant pages. This dimensional shift eliminates both oversubscription and undersubscription issues.
2Quantity of substance
If advertising is presented based on page subject matter, then advertising may reach users interested in that subject, but effectiveness is low because user interest is not accurately predicted
Solution Approach 1:
The patent implements feedback by monitoring actual user activities (pages visited, time spent, interactions) and using this data to dynamically determine user interests and serve relevant advertising. Instead of relying on static page subject matter assumptions, the system continuously gathers feedback on user behavior and adjusts ad delivery accordingly, significantly improving advertising effectiveness while maintaining broad reach.
Solution Approach 2:
The patent changes the parameters used for advertising targeting from static page characteristics to dynamic user behavior parameters. By measuring and responding to user activities such as pages visited, time spent on pages, and interaction patterns, the system dynamically adjusts which ads are shown to which users, transforming advertising from a page-based to a user-based model that accurately reflects actual user interests.
Data Source
AI summary
A facility for Performing on-line advertising at a subject web site relative to a selected user activity is described. The facility receives a plurality of bids. Each bid is submitted on behalf of an advertiser, and specifies an advertising rate that the advertiser will pay to display an advertising messages to users performing the selected activity with respect to the subject web site. When a request for any web page of the subject web site containing advertising space is received from a user that has previously performed the selected activity, irrespective of the identity of the requested web page, the facility both (1) identifies the received bid specifying the largest advertising rate, and (2) causes an advertising message to be displayed in the advertising space of the requested web page for the advertiser on whose behalf the identified bid was submitted.